Podcasts about LCM

Welcome to our vibrant Worship Services at La Casa de Milagros!We're so happy to invite you to spend Sunday mornings together in devotion. Join us live every Sunday at 11 AM for a joyful celebration of God's Love through music, prayer, miracle sharing, dance, and heartfelt joining — a space to open the heart, remember the Presence that unites us, and celebrate the Love extended through us all.This beautiful gathering with Peter, Dini, and the choir continues our deep dive into the theme of "The Gift of Timelessness". This session gently invites us to release our "busy doing" and the ego's exhausting drive for perfectionism, allowing us to rest instead in the grace of God and the inherent perfection of our state of mind. Through inspiring readings such as "No Journey" and moving musical offerings like "Love is Seeking Me" and "Another Miracle," we practice handing over our past and future to fully receive the miracles waiting for us in the present moment.Join Us Live!We are so happy to invite you to spend Sunday mornings together in devotion.  When: Join us live every Sunday at 11 AM.  Where: The Worship Service is streamed online through the LCM app.  Download the LCM App hereBe sure to check your push notification settings to receive a reminder so you never miss a joyful celebration of God's Love. See you there!  Recorded on July 26, 2026,Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

Welcome to our vibrant Worship Services!  This intimate gathering at La Casa de Milagros invites us into a deep and mystical exploration of "The Gift of Timelessness".  This heartwarming episode explores what happens when we loosen our attachments to linear time and trust the Holy Spirit to handle life's logistics. Through vulnerable miracle sharing and beautiful musical reflections, we are reminded that our only true responsibility on this awakening journey is our own willingness.The session also features a joyous announcement regarding the community's Publications Ministry transitioning into a pure "gifting ministry," releasing logistical burdens by offering all books and music completely for free. Woven together with deep readings and acoustic music, this gathering is a profound invitation to step off the timeline, rest in the present moment, and embrace a life of gratitude.  Join Us Live!We are so happy to invite you to spend Sunday mornings together in devotion.  When: Join us live every Sunday at 11 AM.  Where: The Worship Service is streamed online through the LCM app.  Download the LCM App hereBe sure to check your push notification settings to receive a reminder so you never miss a joyful celebration of God's Love. See you there!  Recorded on July 19, 2026,Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

Welcome to our vibrant Worship Services, back this summer from the beautiful, lush garden of La Casa de Milagros! Recorded live on July 12, 2026, this heartfelt session invites us into a deep exploration of the profound truth that "Only Love Remains."Throughout this beautiful gathering, we dive into moving personal experiences of navigating intense grief, the transformative power of spontaneous joy, and the immense comfort found in yielding our personal plans to the Holy Spirit. Woven together with uplifting devotional music and profound reflections on God's sustaining grace—even in the most extreme circumstances—this episode serves as a gentle reminder that we do not have to figure out how to love or fix our relationships. Instead, we can simply allow the quiet, steady voice of the Spirit to lead the way."Only the love of God will protect you in all circumstances. It will lift you out of every trial and raise you high above all the perceived dangers of this world into a climate of perfect peace and safety."Join Us Live! We are so happy to invite you to spend Sunday mornings together in devotion.When: Join us live every Sunday at 11 AM. Where: The Worship Service is streamed online through the LCM app. Be sure to check your push notification settings in the app to receive a reminder so you never miss a joyful celebration of God's Love.Download the LCM App hereSee you there!Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

Welcome to another joyful episode of Sunday Worship, recorded live from the lush, beautiful garden of La Casa de Milagros! In this heartfelt session from July 5, 2026, we invite you to open your heart, remember the Presence that unites us, and celebrate the Love extended through us all.The gathering opens with deep gratitude for the gift of music, exploring how rewriting popular songs with A Course in Miracles lyrics helps let go of the ego and unites everyone in the holy instant.Drawing from Lesson 182, this episode emphasizes finding the eternal innocence of the "Christ child" within us, encouraging us to drop our defenses and let the Spirit lead us to a place of true safety and vastness.Through the lens of Lessons 315 and 316, the presenters share how every interaction with a brother is a divine gift that belongs to us all, helping to dissolve false humility and fear.A vulnerable and moving miracle sharing highlights a personal experience with a severe panic attack, demonstrating how the simple, unprompted smile of a waiter acted as an immediate intervention of pure safety and love from the Spirit.Join Us Live! We are so happy to invite you to spend Sunday mornings together in devotion.When: Join us live every Sunday at 11 AM. Where: The Worship Service is streamed online through the LCM app. Be sure to check your push notification settings in the app to receive a reminder so you never miss a joyful celebration of God's Love.Download the LCM App hereSee you there!Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

The Path of Spontaneity and TrustHosts: Peter & DiniJoin Peter, Dini, and their spiritual community for a heartfelt and entirely spontaneous Sunday gathering. In this profoundly reflective episode, the group explores the absolute peace that washes over us when we set our personal agendas aside and allow the Spirit to guide our every step.Through organic musical performances, deeply vulnerable personal stories, and even a spiritual analysis of pop culture, this episode is a beautiful invitation to look past the ego's projections and recognize the perfect holiness within ourselves and everyone we meet.

https://www.the-christ.net/In this joyful summer worship session from June 21, 2026, we gather in devotion to open our hearts, remember the Presence that unites us, and celebrate God's Love.Join Peter, Jiska, Erik, and Dini as they share profound miracles, deep reflections, and uplifting live music. This week's core theme is Radical Acceptance—exploring what it truly means to accept the Atonement for oneself and willingly let go of the ego's illusions. From navigating the messy parts of our minds to using modern technology for spiritual healing, this episode is a beautiful invitation to surrender to the Spirit.Erik shares a vivid and relatable metaphor about discovering maggots in the kitchen, illustrating how we must allow our hidden, messy thoughts to surface in awareness rather than repressing them.Discover how Jiska creatively uses modern technology, specifically ChatGPT, to navigate the "Levels of Mind" process for deep emotional clearing.The group discusses how feelings of unworthiness and guilt are just constructs of the ego, and how true radical acceptance requires surrendering these beliefs entirely.Enjoy moving, original songs woven throughout the session, including "Jesus Walk With Me," "Radical Acceptance," and "Joy is Here."Join Us Live!Our Worship Services are streamed online every Sunday at 11 AM! Download the LCM app and check your push notification settings to receive a reminder so you never miss a joyful celebration.Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

Isabella Piratininga, diretora de tecnologia e inovação do iFood, explica como o AILO funciona por dentro, o assistente de IA generativa que permite fazer pedidos em linguagem natural pelo WhatsApp e como o iFood constrói inteligência artificial em escala desde 2018.Nesse episódio:- Como o LCM (Large Consumer Model) personaliza pedidos e é 60x mais barato que modelos internacionais- Por que o AILO foi reescrito do zero no AILO 2.0 e o que mudou no framework de decisão autônoma- Como o iFood orquestra múltiplos agentes de IA- Por que 300 deploys em 4 semanas viraram rotina e o que isso muda no trabalho do PM- O erro que a maioria das empresas comete ao implementar IA (plugar API sem governança de dados)- Como o iFood controla custo de IA com seleção de modelo por tarefa e dashboard de eficiênciaIsabela tem passagem por OLX e BanQlub antes de chegar ao iFood, onde lidera o time de inovação e produtos disruptivos. Ela conta como saiu do design, passou por todas as caixinhas de produto, e hoje reescreve o produto mais inovador do Brasil do zero.Se você trabalha com produto digital, IA aplicada, ou quer entender como uma empresa de 8 mil pessoas realmente escala inteligência artificial, esse episódio é referência.

Join Dini, Melly, Colin, Rafael, and Erik as they gather to explore a profound spiritual theme: how giving is the true way out of darkness and any sense of lack. Through personal parables and heartfelt reflections, the group discusses the continuous invitation from their community to step fully into a life of giving.  Colin shares a powerful story of being detained at Mexican customs during the COVID-19 pandemic, describing how stepping up to translate for other travelers completely dissolved his intense fear.  Rafael discusses how fully accepting new community projects—such as organizing Portuguese and French translations—brought him immense lightness and happiness. He also shares how leaning into divine support helped him overcome his feelings of shyness and indignity.  Melly reflects on her journey of letting go of worldly desires to instead ask Jesus to teach her how to give. She shares a touching story about realizing that true change and healing must happen within her own mind, rather than expecting others to change.  Erik introduces his original song, "Giving Trust," sharing his personal realization that letting go of his own point of view to completely trust his brothers is a profound form of giving.  The group shares live, heartfelt musical performances. Listeners will be treated to beautiful lyrics, including reminders like "Whatever appears to be lacking is what I have failed to give" and "Love is not obtained, love is given".  Tune in to discover how shifting from a mindset of getting to a mindset of giving allows divine providence to flow freely, providing exactly what is most helpful for our minds.  Join Us Live!We invite you to spend your Sunday mornings together with us in devotion. Join us live every Sunday at 11 AM (CDT) for a space to open your heart and celebrate the light in one another.The Worship Service is streamed online through the LCM app. Be sure to check your push notification settings to receive a reminder for our next gathering. See you there!https://livingchurchministries.org/media/hbf3gbp/sunday-worship-from-the-garden-of-la-casa-the-gift-of-relationshipRecorded Live: June 14, 2026Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

Welcome back to the vibrant Worship Services at La Casa de Milagros! Recorded live from our beautiful, lush garden, this session continues our joyful celebration of God's Love through music, heartfelt joining, and shared devotion.In this gathering, Erik Archbold, Jiska Stroes, Peter Kirk, and the choir lead us as we open our hearts to explore the profound theme of Resting in the Guidance Together. Through uplifting melodies, shared stillness, and reflective prayer, we are reminded that our connections with one another are sacred opportunities to remember the Presence that unites us, celebrate the light within, and extend Love to all.Join Us Live!We invite you to spend your Sunday mornings together with us in devotion. Join us live every Sunday at 11 AM (CDT) for a space to open your heart and celebrate the light in one another.The Worship Service is streamed online through the LCM app. Be sure to check your push notification settings to receive a reminder for our next gathering. See you there!https://livingchurchministries.org/media/hbf3gbp/sunday-worship-from-the-garden-of-la-casa-the-gift-of-relationshipRecorded Live: June 7, 2026Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

Welcome back to the vibrant Worship Services at La Casa de Milagros! Recorded live from our beautiful, lush garden, this session continues our joyful celebration of God's Love through music, heartfelt joining, and shared devotion.In this gathering, Peter, Colin, Dini, Erik, and the choir lead us as we open our hearts to explore the profound theme of The Gift of Relationship. Through uplifting melodies and reflective prayer, we are reminded that our connections with one another are sacred opportunities to remember the Presence that unites us, celebrate the light within, and extend Love to all.Join Us Live!We invite you to spend your Sunday mornings together with us in devotion. Join us live every Sunday at 11 AM for a space to open your heart and celebrate the light in one another.The Worship Service is streamed online through the LCM app. Be sure to check your push notification settings to receive a reminder for our next gathering. See you there!https://livingchurchministries.org/media/hbf3gbp/sunday-worship-from-the-garden-of-la-casa-the-gift-of-relationshipRecorded Live: May 31, 2026Follow us on:YouTube:https://www.youtube.com/DavidHoffmeister https://www.youtube.com/@LivingMiraclesFacebook:https://www.facebook.com/ACIM.ACourseInMiracles Learn more about David & Living Miracles:https://circle.livingmiraclescenter.org/eventsLearn more about A Course in Miracles:https://ACIM.bizDavid's Spanish YouTube Channel is: https://www.youtube.com/channel/UCP9Gw00CldPUmiu43y7fdWwDavid's Portuguese YouTube Channel is:https://www.youtube.com/@davidhoffmeisterucem

We're announcing AIEWF speakers this week! Take the AI Engineering Survey!Today's guest Ethan first joined us for the LS Paper Club as the lead on NVIDIA Cosmos World Model, but then joined xAI and built Grok Imagine in 3 months:He comes back on Latent Space with some nuclear hot takes: that Video Models primarily get their intelligence from LLMs, not from training on video data, and that the next frontier for truly interactive, realtime, long-horizon world models is to work on LLMs (perhaps Interaction Models as well…)Put it this way: In the near term, the next Sora won't be a better video model, but a video agent.Generative Media may more closely follow the evolution of AI coding which went from focusing on one-shot output performance and cost, to multiturn reasoning and planning models for agents and systems that can plan, edit, test, debug, and submit PRs.At a certain point, coding models got so good that the only significant next step to improve performance was handling the orchestration of these models.Now as the performance of video models increases significantly across realism, consistency, & prompt adherence while becoming more cost efficient, the next evolution of video generation may also be systems that can plan, generate, edit, critique, and iterate across an entire creative task. In this episode, Ethan joins swyx and Vibhu to unpack what it actually takes to build frontier image and video systems: data, VAEs, diffusion transformers, audio-video alignment, inference speedups, and the hidden cost of storing and moving massive video datasets. From building NVIDIA's Cosmos world model to joining xAI as Grok Imagine was being built from zero to one, Ethan He has been at the center of some of the most important work in video generation, multimodal models, and real-time world models.We go deep on Grok Imagine, how a small xAI team shipped its first multimodal video model in three months, why iteration speed matters more than almost anything in model development, and why many of the biggest gains come from fixing tiny bugs in data and training pipelines. Flipbook: The future of VideomaxxingVideo agents are almost a sure bet to be the trend in the coming year. We end with a glance at what's beyond video agents:Flipbook caused a minor sensation this year when it was released, but most treat it as a fun demo. Ethan takes it very seriously — with the speed and cost of inference coming down every year, the future of custom video JIT UI is closer than you think. We talked about why videogen models may become the front end of AI, how generative UI could replace traditional HTML/CSS, why world models need to be real-time, interactive, and long-horizon, and why the future of video generation may depend more on language models and agents than on diffusion alone.We discuss:* Why fast iteration mattered more than meetings* Why small training bugs can drive huge model quality gains* Why coding models may make compute the bottleneck again* How image and video models are trained with synthetic captions* The role of VAEs and latent space in frontier video models* Why image models are the foundation for video models* The tradeoff between temporal compression and real-time interactivity* Flipbook, Neural OS, and the future of generative UI* Why future interfaces may go from user intent to pixels* The hidden cost of training video models: storage, egress, and GPU hours* How step distillation and consistency models (like OpenAI sCM) makes video inference orders of magnitude faster* Grok Imagine 0.9 and large-scale audio-video generation* Why audio-video alignment is harder than text-video alignment* Ethan's definition of world models* Reference-to-video, video extension, and long-context video generation* Why xAI's research communication undersells Grok Imagine* How xAI culture shaped the speed of development* AI watermarking, SynthID, and detecting generated media* Why prompt rewriting matters for video models* Grok Imagine Agent and the rise of video agents* Why language models may unlock better video generation* Robotics, physical AI, and embodied world models* Why Ethan left xAI and shifted focus toward LLMs* Self-managed context, memory, and the next frontier for language modelsEthan He* LinkedIn: https://www.linkedin.com/in/ethanhe42* X: https://x.com/EthanHe_42Timestamps00:00:00 Introduction00:01:25 From NVIDIA Cosmos to xAI00:03:24 Building Grok Imagine from Zero to One00:10:07 How Image and Video Models Are Trained00:18:53 Video Compression, VAEs, and Real-Time Tradeoffs00:22:10 Generative UI, Flipbook, and Neural OS00:32:10 The Cost of Training Large Video Models00:37:04 Distillation, GANs, and Fast Video Inference00:41:21 Audio-Video Generation and Grok Imagine 0.900:48:34 What Makes a World Model?00:55:51 Reference Videos, Long Context, and Video Memory01:00:11 xAI Culture, Research, and First-Principles Building01:09:45 AI Safety, Watermarking, and Prompt Rewriting01:13:10 Video Agents and AI-Assisted Creation01:27:32 Why Language Models Unlock Better Video01:31:15 Robotics, Physical AI, and Embodied World Models01:32:38 Why Ethan Left xAI01:34:16 Self-Managed Context and the Future of LLMs01:38:43 Ethan's Career Path and Closing ThoughtsTranscriptIntroduction: Ethan He, Latent Space, and the Path to xAISwyx [00:00:00]: We're here in the studio with Ethan He, most recently of xAI. Welcome.Ethan [00:00:10]: Thank you. Glad being here.Swyx [00:00:11]: We're also here with Vibhu. you were first coming to us or joining the latent space world because you were working on Kosmos at NVIDIA, and you did a paper. We loved it. you presented it as well, so thank you for doing that.Ethan [00:00:23]: I've actually, I also presented the MoEs twice at latent space.Swyx [00:00:29]: How did you actually hear about us? Did we reach out to you? Is that how it worked?Ethan [00:00:33]: No, actually, I-- the community. Like I realized, oh, there is this online community that people talk about AI and also learn from each other through papers every week through the Paperclip. It's very nice.Ethan [00:00:49]: I learned a lot.Swyx [00:00:49]: I think three years stop. We haven't stopped even on Christmas and New Years. many weeks I want to stop but it keeps going.Vibhu [00:00:58]: No, that was good. I think you had posted that you worked on a paper, and I was “Oh, very cool. We have Paperclip. Present then.”Vibhu [00:01:04]: But I might have reached out to you after.Swyx [00:01:05]: you-- because it's an amateur club, right?Swyx [00:01:08]: so it's very unusual and but we have sometimes paper authors come by and actually explain the paper. Today we just did, the poolside paper, which was apparently very good.Vibhu [00:01:18]: Came out yesterday.Vibhu [00:01:19]: pretty interesting, right? Fully open. They talk about everything, systems. So it's a good one. We'll, we'll recommend people to read it.Swyx [00:01:25]: Bring us up to speed on your transition to xAI, ‘cause I actually don't even know when you joined. just like tell the, tell the story about the sort of transition.From NVIDIA Cosmos to xAI: Scaling Video and World ModelsEthan [00:01:34]: Before xAI, I was working on Kosmos world model as in-- at NVIDIA. So Kosmos is, it's a giant video foundation models that can-- that aims to simulate the world and for-- it serves as a foundation of-- for all of the roboticists to build on top of. There, once I built the Kosmos one, I realized as this thing also has a scaling law similar to language model, we need to scale up the video models further. that's, that's why I realized I need to move to somewhere with much more compute resources. That's how ISwyx [00:02:13]: Than NVIDIA?Vibhu [00:02:14]: The GPU rich came themselves.Vibhu [00:02:19]: And timeline-wise, when was Kosmo? It was pretty early, right? It was open world model, open paper, everything.Ethan [00:02:25]: It was end of twenty-four.Vibhu [00:02:28]: End of twenty-four.Ethan [00:02:30]: Then at mid twenty-five, I moved to xAI. At that time-- I joined about the time when xAI was about to build video models and in multi-model models. There were no infra, no data, and no model, and it just-- as a few engineers, we built it in three months and released the first model, Grok Imagine zero point nine.Ethan [00:02:55]: And since then, I keep working on video models and move more from training and to post-training of the video models. For example, like a reference to videos, kind of like the cameo feature and, video extensions. And, before I left, I worked on a world model, leading a small team to focus on the real-time long horizon video generation.Building Grok Imagine From Scratch in Three MonthsSwyx [00:03:24]: Can you give like a rough roadmap of okay, you're on a brand-new team. Grok previously was only text, or they partnered with BFL for their image gen stuff. What do you-- what are the building blocks, right? You have compute, data you can procure somewhere. Like just what are like the sequence of things that people should think about when you're setting up a new team?Vibhu [00:03:43]: actually even deeper, not just data you can procure. You guys had to go through getting the data too, right? So you shipped it pretty fast, but yeahSwyx [00:03:51]: three months is likeVibhu [00:03:52]: From everythingSwyx [00:03:52]: actually like very surprisingly fast.Ethan [00:03:55]: One thing I say like thanks to my experience at NVIDIA, ‘cause first time when we were building Kosmos together, we built it, for about a year. So this is like the second time I do it. Roughly have an idea, what to do. I say the most important thing is the talent. Everyone were very strong and clever, very close with each other towards a common goal. So that speed up things a lot. So you reduce the communication bandwidth among people, and everyone can work towards the same goal. It's, it's like every day there's not that much meetings on the calendar, like maybe like a, like a sync a day, and after that it's, it's just all building. It was pretty fun at that time.Ethan [00:04:47]: And another thing is that xAI has very strong foundations of like data inference, model inference, and the supporting there can help the model develop a lot. When I look at, training models, I don't so actually the top important thing is like how many, how many iterations can you do, per day? and the more iteration can you do, you can, you can train the model much faster. So if you have very strong infra and you have a lot of compute, you can, you can train these models in very short period of time. That can give you a much larger buffer to, for errors, and it also gives you the opportunity to spot more bugs.Iteration Speed, Compute, and Debugging Model PipelinesSwyx [00:05:46]: What is an iteration? Is it like a few hundred steps or what are youEthan [00:05:50]: Let's say just the train-training the model, like from acquire new data and maybe design new algorithms and train a new model, maybe at smaller scale orSwyx [00:06:01]: So cycle time for like any hyperparam that you're searching.Ethan [00:06:04]: Cycle time and tune to like eval this model. Is this model better than my previous iteration?Ethan [00:06:11]: SoSwyx [00:06:11]: So it's like before you, someone had already set this up that you can iterate very quickly.Ethan [00:06:15]: I think the foundation there is extremely good forDeveloping and research models.Ethan [00:06:23]: And often I find is it-- this is kind of boring, but like a lot of the improvements does not come from new algorithms. It comes from finding small bugs here and there in the data pipeline, in the, in the model training pipeline. Those give, those give the biggest boost to the model quality.Vibhu [00:06:46]: It's interesting, right? So you say it's like small team, less communication bandwidth, but also a lot of quality is like find little bugs. It seems counterintuitive, right? You have a lot of people, you can iron out more of those, but it's interesting to see the other side, right?Swyx [00:07:00]: I also wonder, have you-- do you try using LLMs to look for bugs? I don't know.Ethan [00:07:05]: I remember at that time it was mid two thousand and twenty-five, so it's the coding model wasn't quite there yet. I remem- I remember like December two thousand and twenty-five, it was extremely good. Yeah, I've been, I've been using it at that time. It's, it's helpful. sometimes it produce codes that are kind of difficult to maintain, even though like the first time it built something extremely fast. But it gave the, like a spaghetti code, thousands of lines that I couldn't maintain, and the LLM itself couldn't figure out what's, what's wrong and how to improve on top of it. But now I find it much better. Yeah, I want to bring up another point here is now coding models are much more efficient and can help us implement stuff much faster. Compute might become a bottleneck again because previously, like if you want to train a new model, say you want to generate new synthetic data and then or write a new algorithm, it might take a few weeks. And during that period of time, you don't-- you might not have experiments to run. But now you can build that thing within a few hours, then you can immediately train a model.Ethan [00:08:24]: Now you have to have enough compute to try all of the ideas. So compute might be the bottleneck of iterating speed again.Swyx [00:08:36]: yeah, I actually, honestly, I think it's like kind of a stressful job because you're “Well, I should be trying everything, and if I'm not, then I'm not doing my job well.”Vibhu [00:08:48]: there's also the stress of you're eating thousands of GPUs per hour, which is very expensive and, compute can go to other researchers.Swyx [00:08:56]: You got the daddy Elon toVibhu [00:08:57]: You got daddy Elon.Ethan [00:08:59]: It wasVibhu [00:09:00]: But there's still finite amount of compute, like you want to use it, you want to use it well, you want more of it.Ethan [00:09:06]: That was quite stressful indeed. Yeah, I think one thing is the-- with coding models now, like a lot of these jobs can be automated, which is much better. A second, it's a, it's a marathon, so you got to maintain good health and, a regular schedule.Vibhu [00:09:28]: It's, it's hard to hear that when you shift from zero to nothing in two months.Swyx [00:09:32]: and, I think obviously the culture at xAI is very famously, people work very hard. one thing I did want to dive into, in our-- in the notes that you, that you sent ahead of time, you had specific comments about the cost of Video Gen training. presumably this is on the Colossus-1, right? the two hundred megawatt cluster. Any whatever you want to just share on that.Vibhu [00:09:54]: I think there's, there's three things we're talking about, right? So there's Video Gen, there's also the Image Gen model that you put out. Do you want to like complete the, okay, so zero to one, you have a few months. Just what are the stages of create Image Gen model?Swyx [00:10:06]: Oh, yeah, maybe I got distracted.How Image and Video Models Are Trained: Synthetic Captions, Tokenizers, and VAEsVibhu [00:10:07]: Sorry. and then, from there's Video Gen, there's Audio Gen. Would love to get into those next. But what is that first few months like? So small team, a lot of bugs, iterations, but what does it look like? Do we take something off the shelf? Do we just get data compute? What's, what's the few months like? How do you go to state-art Image Gen model? How do you just start?Ethan [00:10:28]: I cannot comment specifically how xAI did, but it's, it's a quite standard process. I can draw some, examples from Cosmos. So mainly it's building a video model, you actually need to build a image model first. And building these two models, the data you need is a hundred percent synthetic pair of language and image or language to video. Because on the, on the internet, actually, the videos don't naturally associate with text. So you can say, oh, like on YouTube, you have the title and you have the description and the commentsSwyx [00:11:11]: TitleEthan [00:11:11]: of a video, but usually they're not relevant to the video itself. And say maybe like the video is a natural scene of mountains or something, and the title is, I'm so happy today.Ethan [00:11:26]: So they have they have no correlation at all. So the first step is to, you have to generate synthetic pair of language with the videos. So you gather videos from the internet, and you use a VLM to caption the videos. So that part, here's a question, like how do you, how do you gather VLM to begin with? So if there's noSwyx [00:11:55]: You, so you fuse the model, right? LikeEthan [00:11:57]: Say if there's no like VLM exists, like how do you generate the text to the beginning, right? It's, it's impossible.Swyx [00:12:04]: I see.Ethan [00:12:05]: In the beginning, it's like you ask human to describe the video as detailed as possible.For example, you ask them to describe everything, like all objects, all characters, and all interaction and dialogues in the, in the videos. So that's in the protocol of Cosmos labeling. We require the objective we give to the labelers was that you have to describe the video as detailed as possible, such that a blind person hears a blob of text can reconstruct what the video is like from their head.Swyx [00:12:43]: Video or image? You're talking about images.Ethan [00:12:44]: Video or image, either one of them.Vibhu [00:12:47]: This was pretty common when we went from clip and DALL-E, right?Vibhu [00:12:51]: It's all training on really detailed captioning of images. So same is applied to video, but insteadEthan [00:12:57]: same appliedVibhu [00:12:57]: of using multimodal model to pass in video images and write rich descriptions, you can alsoSwyx [00:13:04]: I think there's this traditional perspective of supervised, or, very highly human curated thing. I feel like there's a unlock with unsupervised, right? Where like you have enough to bootstrap that you can just throw common corpus on it or, whatever. like unsupervised vision and language pairing, right? Like where you just have, interspersed image and text and it just learns. To me, that is the VLM breakthrough that is different from the clip, different from the LM era.Ethan [00:13:36]: It's interesting to see that you kind of need both data.Ethan [00:13:41]: For example, for theSwyx [00:13:41]: You need it to bootstrap it up. YeahEthan [00:13:43]: for the generative model training, there's also usually like a small percentage of unlabeled data. So the model is instructed to generate a video without any text instruction. That can also help the model generalize. So after this stage of generative synthetic pair, so, one important common step is to train a compressor or a tokenizer of the image or videos. So because, if you train-- If you can technically, theoretically train image or video models on pure pixels, but the problem is that the, it's, it's a lot of tokens. So like one image, it's, a thousand by a thousand, it's like one million tokens, one million pixels. It's impossible to train transformer on that. So it's, you need to train a tokenizer, which can go from image to latent space and latent space back to image.Swyx [00:14:45]: That's why we named the podcast.Swyx [00:14:48]: But, basically, you're talking about vocabulary science.Ethan [00:14:50]: so vocab.Swyx [00:14:51]: And so, what is, what is imp-- like a million is impossible?Ethan [00:14:54]: In generative models, the vocab is continuous. It's a continuous space. We can think about like you map an image to a vector. It's a, it's a fixed length vector. It's sixteen or forty-eight, something like that. And then you map that vector back to the image space. And the mapping is, has-- The mapping is patch-based. So you say you haveEthan [00:15:22]: a sixteen by sixteen patch and you match, you map that patch of pixels into this latent space.Swyx [00:15:29]: We've covered thisVibhu [00:15:30]: This is like the vision transformersSwyx [00:15:32]: VAEs,Ethan [00:15:33]: VAEs.Vibhu [00:15:34]: You basically compress your input, you do your generation, you're reasoning all that generation in smaller dimension, and then you project back out.Swyx [00:15:43]: VAE is a form compression, but I think the for me, the patching thing is from VIT, right?Ethan [00:15:48]: You can make those.Swyx [00:15:49]: Literally the, yeah, the paper is titled like sixteen by sixteen is all you need. something like that. and then I think also, people make a lot of comparisons with this kind of patching with convolutions.Swyx [00:16:02]: Which is you're, you're kind of re- reconstructing the old paradigm with the new.Ethan [00:16:05]: Actually, in VAEs, there are, there are both convolution networks and transformers. You can actually do both.Ethan [00:16:14]: After this VAE, so what you've got is you've got latent space tokens and you've got the language tokens. So now the training of the diffusion transformer, usually generative models use diffusion transformers. It is actually quite standard. It's, it's very similar to how you train a language transformer models. It's not that much difference. It's just the tokens, the visual tokens in, visual tokens out. The only difference is there's a denoising process. So you train the model to unmask some of the noise. So you add, you add random noise to the visual tokens, and then you train the model to remove those noise to generate the clean tokens. Any inference, the model can iteratively remove noise from a hundred percent noise.Swyx [00:17:12]: And then there's also, to speed things along on the tech tree of diffusion, there's CFG, and then there's, there's also, latent diffusion that, there's, there's someone in there. I think, somewhere along the line, obviously, like stability and all these other guys, pioneered a lot of this, architecture. I don't know if you want to get into that or just, or do the video side up to you.Bootstrapping Video from Image Models and Temporal CompressionEthan [00:17:37]: After you train such model, such image model, the reason it's a, it's a foundation for video models is that image models are cheaper to train, and they have much denser connection between language and text. So, sorry, language and images. For example, you train a billion, you train on a billion images, and there's a mapping from the text to the image. And the cost to train the same, like the, a billion, a billion text to a billion videos, that's much more expensive because videosNaturally have more tokens than images. Because the diffusion models, their understanding of, language purely come from this mapping. So if you don't have enough mapping, so if you only train on like a ten million videos or something, there-- you might not see enough language tokens in your training, so your model does not understand human intention enough. So that's why you really-- you train-- you first train this image diffusion models, and then you bootstrap the video model from there.Swyx [00:18:53]: One thing I did want to ask, because I-- actually, I think you're, you're the first per-- video model person I've ever talked to, I think. we've, we've like talked to Luma and all those folks. There's all these tricks in video compression where basically frame by frame there's not that much difference, so actually you don't have to regenerate or save the whole frame, right? but I think MP4 compression or something else like that.Swyx [00:19:16]: is it tempting to use that? Or as far as I can tell, everyone just treats it as, “No, we would just generate every frame.” Is that roughly the state-art?Ethan [00:19:27]: There are a few different approaches. Let's say first, like you want to just directly use MP4 compression and use that as the tokens for the transformers to train, right? So people actually have tried that, but the main challenge is the latent space for the MP4 tokens were not, were not very comprehensible for the models. It's, it's extremely hard to train on that. And there's aEthan [00:20:01]: So that's why they created VAEs, which creates more continuous, latent space, so the models can understand that latent space and learn from it much easier. Even within the VAEs, there are different difficulties of the latent space. So you can imagine something the simplest, the most naive VAE is like you have an image, and you just shuffle all of the images into a, into a vector. So you don't need to train any VAEs, right? But that latent space is extremely hard for models to train on top of. That's why there are some debate on like how do you compress the tokens. So you mentioned like you can compress frame by frame. Also, you can compress, the temporal dimension.Ethan [00:20:52]: The difference is if you compress the temporal dimension, you get a much higher compression rate. Because there's temporal redundancy between frames, because, this frame and the last frame, likely they are mostly similar, so there's only some small difference. for example, I think in 12.1 VAE, they have like a eight by eight by four compression rate. So the four temporal tokens are compressed into one tokens. That can save a lot of, save a lot of the context length. If you do it frame by frame, you have to do maybe like eight by eight by one. Your context length will be four times larger. That being said, the benefit of the frame-- per frame compression, we might come back to this later, is, real-timeness and interactivity. ‘Cause if you, if you strain the output of the model, frame by frame, you can-- the model can respond to any user request immediately. So if you have like a temporal four compression, four times compression, thenSwyx [00:22:06]: It might be laggyEthan [00:22:07]: there's a lag there in nature.Swyx [00:22:10]: So you're very pilled on this. let's just go ahead and bring it up ‘cause we have the visual prepared anyway. There's some frontier applications of real-time video gen. So Flipbook is one of the examples that went viral recently, right? What is Flipbook?Real-Time Generative UI: Flipbook, Neural OS, and Diffusion Front EndsEthan [00:22:23]: Flipbook is kind of like a web brow- web browser. You can see like it has the web bro- browser UI on top. The difference is all of the UIs are generated by generative image model in real time, and anything here are fake. But you can, you can explore inside this wor- this imaginary world. Say like we-- here we have engineering the Great Pyramid. Like the model generates this for us to understand how it works, and if we want to navigate around and understand further, we can click on some of the, some of the description here, and the model will generate a new page, new subpage describing the details we want to know about.Swyx [00:23:14]: So it's basically kind of we're playing a video, but it's pausing for our next interaction, and then it just plays the next thing based on our interaction.Swyx [00:23:23]: Which is kind of cool.Vibhu [00:23:25]: and you kind of decide your story. So this was, how do you make a pyramid? levering technique seemed interesting, right? It shows how do you take Okay, I want to know what is thisSwyx [00:23:35]: The demo, the demo tweet had more animation between frames.Vibhu [00:23:38]: I think it's just skipping,Swyx [00:23:39]: Oh, it's just skipping a lot of frames.Ethan [00:23:40]: they also have a video modeVibhu [00:23:42]: It takes a lot. There's a lot of peopleEthan [00:23:42]: but, a lot of people are using it.Ethan [00:23:45]: So it's not available.Vibhu [00:23:46]: There's a live video stream. We can try,Swyx [00:23:50]: So this is an example of the kind of future that you see at the extreme. We don't-- we're obviously not in it today.Swyx [00:23:56]: But in a world where inference is completely free this is better than generating code and text?Ethan [00:24:02]: So this is, this is a final state of where Viva will be at for word model, I think. Imagine internet doesn't exist, and then you type in google.com. Like what should, what should, what should a model show you?the model can imagine something, and this is what the model imagine. And these web pages, they completely do not exist. So I think as the inference costs come down, we are going to have generative UI for everything. If you think about how the coding model works, so they write code for a web page, and they render the code might be con- converted into binary, and the binary render the pixels on the screen. So we in machine learning, every time we have some breakthrough, obviously it's, it's more intuit. So why don't we have like user instruction to the pixel directly? So the generative UI will be user intention to the pixels directly. And say like even if I want email, let's say everyone have the same interface, but I want, I want it slightly different. I want the email to show to me like a TikTok, so I can swipe left and right for the emails. And or maybe you want something else. We can have completely different things. Or like I have I'm looking at, Instagram stories, and I don't like the Like button. I always may click it. And, generative UI resolved it. So it's going to be a revolutionary replacement of the interface. So in the future, we might have much more powerfulEthan [00:25:50]: LLMs and coding models running behind the scene. And in the, in the front-end, the diffusion model will actually be the front-end to show stuff to you. That's how I imagine it.Swyx [00:26:02]: Diffusion front-end, deterministic back-end.Swyx [00:26:04]: Something like that. I find that very expensive, but,Vibhu [00:26:08]: I find it interesting you called LLMs writing code on the back end deterministic, but okay.Swyx [00:26:14]: you write it onceVibhu [00:26:15]: Compare it toSwyx [00:26:16]: And then you execute.Ethan [00:26:17]: If you think about the cost, say, let's say H100 costs $1 per hour, and if you use this eight hours a day and thirty days, so, every month you're paying this two forty, you'll actually not wanna pay for that. That's even more expensive than Cloud Code Max. But if you think about the compute costs come down like two times every year, and I think the future will likely arrive like within few years.Vibhu [00:26:49]: It's everything, right? compute cost comes down, compute gets faster, model gets smarterEthan [00:26:54]: More efficientVibhu [00:26:54]: model gets smaller.Swyx [00:26:55]: I don't know why you say two times, ‘cause I think it's like 100 times. In language models, it is roughly one hundred to a thousand times every twelve to eighteen months, for the same given level of LMSys, ELO.Vibhu [00:27:08]: That's a net of everything, right? That's model performance alongside compute. So different than just compute costs come down. But, a very interesting future.Swyx [00:27:19]: So the web designers will have to shout out that accessibility is an issue, right? how do you deal with screen readers or whatever. But yes, this is higher bandwidth storytelling than anything you can possibly generate with code, right? So I think that's the rough idea.Ethan [00:27:34]: And I'd like to add a little bit that so human naturally have the maximum bandwidth when we are looking at things, look at videos, and we also have maximum output bandwidth when we are talking. So in the future, it might be something like we talk to AI models, and the AI model responds back with a generative UI. So that would be the maximum input and output bandwidth to interact with AI models before neural link happens.Vibhu [00:28:06]: And it's also very custom, right? Some people are very visual, some people are not as visual, right? They prefer the text. But the best thing about generative UI, right, it can also be text.Swyx [00:28:17]: There's another project that we wanted to highlight, which is the Neural OS. Kinda similar idea, but here you're literally operating, simulating an operating system with a video model.Swyx [00:28:27]: and you can play Doom, you can do Firefox. I find this like mildly less impressive, obviously, because it's an OS that I can run.Swyx [00:28:37]: But here everything is imagined.Vibhu [00:28:40]: I was, used to the Command+W to close the Firefox tab. It didn't crash. That's why I saidSwyx [00:28:45]: It's too immersive.Vibhu [00:28:46]: It's, it's too immersive for me.Swyx [00:28:47]: Too immersive.Vibhu [00:28:48]: I wanted to close the tab.Vibhu [00:28:49]: But yes, I can play generated diffusion.Swyx [00:28:51]: this is shockingly fast.Swyx [00:28:54]: Because I remember there was a demo about like maybe one to two years ago. Someone tried to do the first-person shooter with a image model. There was no consistency. It was very slow. But here it looks like realistically it's-- this is Doom.Vibhu [00:29:07]: I think there's two sides to that, right? There's okay, what is running a game? The heavy part of it is actually the game engine, all the lighting, all that stuff, the graphics. This is just kind of video, right? Like we've solved consistency. This is still, it looks like a few years old image generation. There's some temporal consistency, but it's, it's kind of just images stitched together as frame video. But it's a good visual representation to pi- to picture the future you wanna see, right? that's, that's what I see in these more so.Ethan [00:29:38]: This reminds me of how the video models gets better and better. So Neural OS is kinda if you just look at it feels like it's just a crappy version of the, like the Windows we could have, right? And, but the difference is, so the model, this model is overfitted on the existing operating systems. It can generate nothing different than that. But it's actually also similar to video models. So when we are training these video model, image model, we train them on internet. There's no imaginary supernatural stuff on the internet. But once we train this model, you can prompt the model to generate something supernatural that have never existed in the data set. So if you train your Neural OS or neural computer on the standard screen recordings on the entire internet. The model can imagine completely new interface to interact with the computer.Swyx [00:30:43]: This is one of those things that is magical to me. usually generalizing out of distribution is bad, but somehow we have learned some kind of internal world model that you say, this plus, but it looks like rainbows and butterflies, it'll do it and it will kind of make sense.Swyx [00:31:03]: So yeah, that's kind of cool. Yeah, I don't know if there's any comment more on there. I do, I do wanted to, I did wanted to touch a little bit more on the model architecture stuff, which I think you were getting. It's, really fascinating. We don't get a chance to talk about this enough. So one of the papers that we covered, we've covered every annual, segment anything release. and I don't know if you follow-- you're a computer vision guy, so youEthan [00:31:26]: I knowSwyx [00:31:27]: . So they did memory attention, which is kind of interesting. And I always think, anything where you can, across the temporal dimension, keep some consistency, I think it's, very fascinating, and I don't know if Basically, does that-- the CV side bleeding into video gen side, I think is underexplored, right? we talk about it for labeling, but actually you can borrow the architecture itself.Ethan [00:31:50]: There's, there's also complete different approaches, right? you brought up the term world model, so we went from video model to world model. There is diffusion, but there's also other approaches that people are doing. So maybe we get into those after as well,?Swyx [00:32:03]: He has a whole definition of world models and stuff. I feel like we threw a lot at you. Whatever you want to comment on.Why Video Models Are Expensive: Storage, I/O, and Training ScaleEthan [00:32:10]: I think one thing that we should actually comment back on is okay, so we were talking about the steps to train image gen to video model. One thing we don't see as much of is okay, you brought up the delta in training data, right? SoEthan [00:32:24]: you won't have as much a video model might not generalize, but what is the cost of training a large video model? So we know for LLMs roughly, okay, even like the poolside thing that came out today, right? It's a Gemma level model trained on roughly forty trillion tokens at this many H200s over this much time, right? You can see what is the exact cost of that. So how many GPU hours over how much H200 costs? So how do we do the back-end math of, same thing for video models, image models. How do you, how do you kind of break that down? I can share some back-envelope calculation. So surprisingly, video models is-- the cost is very-- is comparable to language models and obviously the largest scale is language model, maybe like a medium scale to language models. I said just storing the videos alone, it costs a lot. You can, you can maybe look up on AWS or something.Ethan [00:33:20]: You really, say if you have a billion videos and let's say, let's just say like each video, like five megabyte, then you need five petabyte to just store those videos. And also remember we talk about you use a VAE to compress the videos, and you also need to store, typically you need to store those continuous feature, in-- also in your storage. That's also comparable size with the videos themselves. So just storing these videos and the features is tens of petabytes alone. And,Swyx [00:33:58]: I just, I just looked up the calculation. Five petabytes on S3 Standard is one hundred K per month.Ethan [00:34:05]: AndSwyx [00:34:05]: It's comparableEthan [00:34:05]: and you needSwyx [00:34:06]: AndEthan [00:34:06]: And then like tens of petabytes, two hundred K. And even more expensive is you have the ingress and egress.Swyx [00:34:13]: Oh, yeah.Ethan [00:34:14]: Like you-- through the internet. You have to just to download those videos, I believe it's, it's more expensive on AWS than just storing those videos.Swyx [00:34:25]: Storing, yeah.Ethan [00:34:25]: And each training runs, you probably need to pull them once. If you train multiple times, it's, it's even more than that. So it's like just storing the network, those costs is just, it would be a few, a few millions per month to just storing everything, not to mention the GPU cost.Ethan [00:34:45]: AndSwyx [00:34:45]: my side tangent, the compute rental, like GPU rental is very efficient. There's one side, okay, you can be XAI and build your data center. Should we not just build our, storage compute as well? LikeEthan [00:34:57]: Of courseSwyx [00:34:57]: cloud cost compared to just,Ethan [00:34:59]: You save so muchSwyx [00:35:00]: store. Yeah, exactly.Swyx [00:35:01]: Especially with like egress and stuff. So.Ethan [00:35:04]: That's a good idea, but it also comes to-- there are some of its own challenges.Swyx [00:35:09]: Of course, of course.Ethan [00:35:10]: like people who build the GPU data centers, they might not expect this much, storage. And yeah, people build storage, typically they just build it somewhere with just CPUs.Swyx [00:35:23]: I just looked it up. Five-- AWS only charges for egress, not ingress. Tier five for five petabytes is two hundred and thirty K.Ethan [00:35:32]: Even more expensive than the storage.Swyx [00:35:34]: But storing is per month, right? You check in, then you cannot check out. so it's so cool. It's okay. So there's that side.Ethan [00:35:41]: So the TLDR, my backhand mathSwyx [00:35:42]: Data is larger than you think. Yes.Ethan [00:35:44]: my backhand math of GPU hours times GPU cost is also very much, I'm missing some storage.Swyx [00:35:49]: You're also-- you're basically like also more IO bound than normal training.Swyx [00:35:55]: Yes. ‘Cause like data loading, so caching everything, it becomes super important.Ethan [00:36:00]: So in Cosmos, we did a lot of optimizations to make it not IO bound. So, speaking of the training, actually training the model, the GPU cost, if you look up like the open source model, how big these video models are, I think like LTX has nineteen B parameters. That's a dense model. And people are also exploring, MoEs, so it might be twenty B active and, like a hun- hundreds B, total. So that's, that's even-- that's similar size as medium-sized LLM models. And if you, if you look at number of tokens-Uh, we disclose that in Cosmos. It's also like tens of trillions of tokens on the visual tokens. So putting this together, the cost of, training these video models, it's actually comparable with LLMs. Not to mention, the infra is slightly different from LLM, so it might be less efficient to train these models.Inference Speedups: Step Distillation, Consistency Models, and GANsSwyx [00:37:04]: Do you get the benefits of traditional diffusion speed-up? So for, images, there's LCM, LoRAs for, fine-tuning. There's, there's a lot of stuff that's beenEthan [00:37:15]: Flow matching.Swyx [00:37:16]: there's flow matching. There's a lot of stuff that's been done. there's some overlap that applies to diffusion on the inference side and stuff or?Ethan [00:37:23]: so the difference-- the inference side is a completely different story.Ethan [00:37:28]: I think for the training side, it might be a little bit hard to reduce that cost. And for the inference side, the biggest gain is from the distillation of these models. You can-- It's called step distillation, slightly different from knowledge distillation in LLMs. So you-- Typically, for flow matching models, you need like 100 steps or something. Like a distortion model even need even more, like 1,000 steps to generate a good image or video. A step distillation is try to learn to generate fewer step from the model itself. It's kind of like now we-- you use the full model to generate in 100 steps, and then you take a model that only generate 10 steps and let that model to learn from the perfect one.Ethan [00:38:25]: why this workSwyx [00:38:27]: Strong to weak seemingly.Ethan [00:38:28]: It is. It's kind ofSwyx [00:38:29]: DistillationEthan [00:38:29]: kind of like strong to weak. the-- from the modeling perspective, the strong model, the teacher model is trying to model the image and videos of inter-internet, and that distribution is extremely complex. But the step distilled model is just trying to learn from the teacher. The teacher is a model, and the size is fixed, as the distribution is much simpler than the whole internet. That's the intuition I have why step distillation can work. So usually these models serve in productions, they only run in a few steps. In Cosmos, I believe we have, we have like four step and eight steps. If you do some simpler task, image-image translation, it can even run in fewer step, like one step in Cosmos Transfer.Swyx [00:39:22]: I think this is the same intuition that guides a lot of the consistency model work. I sent you a link for, SCM. I don't know if you covered that. To me, that was actually one of, the most impressive papers I've ever seen from OpenAI.Swyx [00:39:34]: That this is the unifying grand concept of consistency models. I don't know if you have any comments on this.Ethan [00:39:41]: So there are, there are a few different approaches,Swyx [00:39:46]: Oh, yeah. Here it is.Swyx [00:39:47]: Two steps versus twenty or 100 steps, whatever. It's already done.Ethan [00:39:52]: So there are, there are a few different approaches, for example, consistency model, and there are also Actually, we shouldn't forget GAN. So GAN, actually, that was, that was the OG ofSwyx [00:40:05]: OGEthan [00:40:05]: step distillation ‘cause it trained just one step to begin with. So actually, a lot of, uh-- For example, there's a distribution matching distillation which use, which uses GAN, as one of the laws for distillation. It-- GAN just tells you, “Hey, generate an image,” and thenEthan [00:40:31]: it has a discriminator to tell, is this image real or not? So the model, the model just need to learn one of the distribution, not the full distribution. Because in training, the model is asked to reconstruct the ground truth image from the internet, which is extremely hard. And in-- When you're training GAN, it's a step process. It's just a, “Hey, you generate image. Does this image look as real as the image from the internet?” Which is a much simpler task. And, yeah, combining a lot of these approaches together, people typically do that, like consistency model and distribution matching and GAN, and we can get these few step models.Audio-Video Generation and Time AlignmentSwyx [00:41:21]: Then there's one step I wanted to add, which is audio and video.Ethan [00:41:26]: So, Grok Imagine zero point nine, I believe it's, it's a first audio video transmodel deployed at a large scale. SoSwyx [00:41:39]: And that was your first model?Ethan [00:41:40]: that was, Grok Imagine's first model. It's, it's audio video, joint generation. I think the hard part is, the modality alignment, ‘cause before this transmodel, we have, we have text to video alignment. We have this, correspondence between text and video. Typically, most of the VLMs, they understand images and videos. Video's very rare, and they don't understand audio mostly. And if you look at the audio generation on the LLM side, you can talk to them perfectly fine, but if you ask them to sing a song or something, it typically is not very good. Also, they don't have, they don't have music either. The hard part is thatUh, actually audio has two component. It has like a discrete component, a continuous component. The discrete component is like the language.Ethan [00:42:44]: So when we speak, it's just, someSwyx [00:42:47]: It's an ASR issue, yeah.Ethan [00:42:49]: It's, it's text token with some characteristics, I would say.Ethan [00:42:54]: But musicSwyx [00:42:56]: I think the speech guys would disagree with this.Swyx [00:42:57]: Like disfluencies and then,Vibhu [00:43:00]: There's tones you can get angry.Ethan [00:43:01]: Well, I say largely.Ethan [00:43:03]: the mu- but the music is completely different. It's, it's very continuous, and you cannot model them like discrete tokens in language models. this is like the hard part for models is, not to mention we have to align text, video, and audio together.Ethan [00:43:26]: SoVibhu [00:43:26]: How?Ethan [00:43:28]: So significant-- some significant challenges are like-- So first, like we talk about as the VLMs, they cannot understand most of them cannot understand audio.Ethan [00:43:39]: So you have to have some way to do the synthetic data generation for audio. You have to caption the model, and that involve, that involve synthetic data and human data effort a lot. And not just surprisingly, most of the LLMs are very bad at recognizing, like the beat, tone, and the details of the of music. They can, they can give some general prediction of which song is this, but it's very hard to describe the details of the music. like we mentioned in image generation, like you have to describe image as detailed as possible so that someone blind can reconstruct that. So here is like someoneVibhu [00:44:32]: DeafEthan [00:44:32]: someone deaf can reconstruct how the music sounds like without actually listening to it. Maybe you can think of it need to have the-- or they call the script.Vibhu [00:44:49]: Subtitles, yeah.Ethan [00:44:49]: You gotta have all the details of the music, and the dialogue.Vibhu [00:44:55]: So is the challenge there typically stuff like music and audio, or is it just Like is there a baseline? Okay, there's enough data where we can understand, narration, conversation, but there's nuances in audio that's where you hit all the data issues or is it just from stage zero, you just do it all right?Ethan [00:45:15]: So one important thing is like the alignment. So the model, the model has to know like the video and audio, the, uh-- it has to have a time-based alignment, like at which time step the video and the audio token correspond to each other. But we actually don't have this kind of alignment for most of the other modalities. If you think about like text and image, text and video, they are loosely aligned. So you can, you can have a description of what's going on in the video, but you don't have to exactly, You typically don't have exact description, oh, at, time step one second like what happened?Vibhu [00:46:02]: It's veryEthan [00:46:03]: At time step two second what happenedVibhu [00:46:03]: coarse. Yeah.Swyx [00:46:05]: So what was the ideal time step? You have to oblate it, and then it's like four seconds or something.Ethan [00:46:09]: So that comes down to how you design the model to, for the model to be aware of as a time, as a time modality. So the model is like a time aware. And that's something pretty unique if you think about LLMs. So if you ask LLM to complete a task, say they, uh-- you ask them and they will say, “Oh, this task will probably take twelve hours to complete,” and they come back in one hour. Say “I've already spent two days on this and I've exhausted everything.”Ethan [00:46:47]: So the LLMs them-themselves, they don't have a sense of time there.Vibhu [00:46:53]: I actually don't think that's just them not having a sense of time. I think it's somewhat based, right?Vibhu [00:46:58]: Like you tell someone, “Okay, go work on this feature. Go implement this,” there's a general understanding you would have of how long that would take without LLMs working at LLM speed, right? So you think back like two years ago, if I tell you to like build me like a new front end for latent space, have a search bar, have all this, you'll estimate that it'll take a few days, right?Vibhu [00:47:19]: So you tell an LLM, “Go build this.” It'll take me a few days. But I think it's somewhat grounded as opposed to them not having the best-- Not saying that they have a great understanding, but I think that example is like you can see where it comes from, right? You're trained on all over the text.Swyx [00:47:35]: They're, they're trying to estimate what a human would say.Vibhu [00:47:37]: because that's what the, that's what the data kind of represents. It's not themEthan [00:47:41]: It came from the corpus on the internet. People have a estimate of how much time.Vibhu [00:47:45]: And not even just in direct like training samples, right? Just your world understanding of tokens of how long stuff takes, right? Go read a book. It'll take you a while, right?Vibhu [00:47:56]: Even if you do nothing but read a book, it takes a few days. So yeah, LLM, I read it took me a few hours.Vibhu [00:48:01]: It'll take me a few hours to go through this research. But this is a tangent.Swyx [00:48:05]: Somewhat, yeah.Swyx [00:48:06]: This is a train of thought I haven't really expressed until now is, which is basically like a full world model must also be recursive, meaning that the participant in the world model must also be aware that they have a world model. which is like this whole recursive thing down the, down the line. but yes, and that the world model can be wrong and that they need to update it and blah. Yeah. We've, argued this on the, newsletter as well, that there needs to be sort of recursive or adversarial world models.World Models: Real-Time, Long-Horizon, Interactive VideoVibhu [00:48:34]: just, to ask, how do you define world model?Swyx [00:48:38]: Oh, yeah, let's go there.Ethan [00:48:40]: SoVibhu [00:48:40]: So just for context, we talked about, video generation, and then there's a-- if you say there's a distinction between world models, what's your, what's your definition? How do you see the two?Ethan [00:48:53]: So disclaimer, I'm not going to debate, what is world model. Yeah. there are many definitions, so I'll just talk about my definition. Since I came from the multi-model, multi-model domain, so mainly talking from video. So world model is like real-time interactive long horizon videos. So there are three parts. so we-- let's talk about them one by one. So the so interaction, so we just, we just look at Facebook and neural computer. So the interaction part of it, so you, world model can allow you to interact with them through keyboard, mouse, and maybe also voice. So these all is-- all is a modality. You can, you can interact with the model, and the model should respond reasonably. Second part is real time. So once you, once, say, you move your mouse, if, say, the world model generate a game, how fast can the game respond? So if you're like professional CS: GO players- -my say, oh, you have to respond- He's beginner within sub ten milliseconds or- Yeah even less. So that's not most of the- No, sixty FPS. Let's go. Oh, three hundred FPS. Oh, five hundred FPS. Wait. okay, yeah. I didn't do the math, but yeah, okay. Uh- Yeah, three hundred FPS, that's a three millisecond. So you have to respond- Oh, s**t. Okay. YeahEthan [00:50:29]: within a millisecond. Most of the video models cannot do that. Yeah. And, but if you, say, if you have a video model that is, say, like a digital human, the response time might be more generous. Maybe typically, for real-time voice interaction, it's like two hundred millisecond. So that's, that's much more generous. But even two hundred millisecond is pretty, it is pretty tricky, ‘cause remember we mentionedEthan [00:51:01]: you have this, temporal compression coming from the VAE. So if you, if you don't compress the temporal dimension, your sequence length is going to explode. So if you want to have this real-time, real-timeness in your model, you have to do is one context problem. And the third part is long horizon, ‘cause we-- if you're not going to just play with, video games just, a few seconds, most video models only a few seconds. We're going to play with minutes, hours. The model have to be able to generate long-form content.Ethan [00:51:42]: So putting these three together, it's, real-time, long horizon interactive videos. I think the final state will be, for example, like a video, a video version of Playbook, where you can, you can interact with, a neural computer. You move your mouse, and you click on the generative interface, and it will reply to you through pixels- generating in real time. But getting there, it's, it's a very long way to get there. So one of the first step, at Grok Imagine, where I led a small world model team there, was to build video extension. So, video extension- it's the first step of interactivity. Yeah. It's, it's the first step. Yeah. So it's the first step- You have it here, video editing, yeah. Yeah. Yeah. So the first step is because, this unlocks long horizon videos. Typically, for most of the video generation models, you give it a prompt or an image as an initial frame. You generate video, that's it. That's just, one time, done. And some creators would try to, use the last frame as a first frame for the second video. It can-- sometimes it works, but if you do it a few times, it says the quality would decrease. And- It doesn't have that context- Yeah over the full video, so the temporal- Yeah, exactly. Yeah, ‘cause you only gave it the last frame, of course, right? Yeah. Exactly. And- it's actually a pretty fun hack. if you've seen like- Oh, no, he's saying something better. Yeah. And for example, like Vue, I remember Vue 3 has like a second context of the last video. It is slightly better than using the last frame, but it has the same problem-- similar problem that it, the quality would decrease. if you extend a few times to, one minute, the video quality would look much worse than the first video. Second, another problem is that the model doesn't have long-range knowledge of, what's happening before. Say, if they generate some dialogue, some, two people speaking, and their voice might change, over some time, especially if the second conditioning, it does not cover the previous context. So these are the core challenges. So the Grok Imagine video extension, it has historical context of all of the previous generated videos. It can, It has, it has the context of, who is speaking and what objects have appeared and everything, having that to generate the next video. So if we naively do this, you can imagine, just, put all of the previous history video tokens into the context. The context lens will easily explode. Especially for video models, that can be like a few, a few million context, I would imagine- context lens. Yes.Yeah.Swyx [00:54:58]: Let's run with that.Ethan [00:54:59]: for example, like in Cosmos, I think just five seconds of video is like a fifty K or sixty K number of tokens. So like if you do, if you do fifty second, that's a five hundred K tokens. If you do longer than that, easily explode. This long horizon, problem was the first step we're trying to solve world model. It turns out people, yeah, people love video extension. Like a lot, a lot of the creators love using video extension to create longer form videos. This is the part I liked that you have a, you have an intermediate step toward the final goal instead of just a straight shot to the final version very much.Swyx [00:55:48]: But I can see you have a strong vision of where we want to end up.Long Context, Redundancy, and Efficient Interactive VideoVibhu [00:55:51]: Does it seem like it's an efficiency issue? okay, we're at a few million tokens context,. If you draw the parallel to language models, we had very short context, two thousand, eight thousand, then, you scale it up one million, ten million. sure, there's effective context, but at the end of the day, it's just what's it worth? sure, there's a whole training data side. In video, it might be slightly easier ‘cause we have a hundred million token video, right? Just take a movie with the full context there. Like is this efficiency from an inference standpoint that like it's expensive, but we know how to solve it? Or like why is this not the approach? So like my broader point was on your second point of world models, you say it needs to be interactive and live, right? You should be able to play a game and see the interaction live. So one thing I see with research is a lot of what you actually serve is different than what you build, right? So we talked about distillation. You train big model, you distill it, you do quantization, speculative decoding. We do all this stuff to serve it efficiently. Should we not just have a solution, like a world model that can interact well, do inference optimization, serve it, distill it secondary, so make it real time after you solve it? So like a-- another parallel is say, continual learning, right? What we need is someone to solve it and show it works inefficiently. Give it a few years, people will make it efficient. Same thing with regular attention, right? It worked. Over a few years, people have different forms of attention, and we've scaled it to be efficient at log context,? So kind of two things there, right? One is it seems like it works. You've scaled it. Can we not just scale it a lot more efficiently over time? Do we need a separate approach if this works? And same thing with interaction, right? if we can get it done, like if we can solve some way that it works, we can solve making it more efficient from an inference standpoint later.Ethan [00:57:53]: that's actually a very good point. So in videos, there's actually a lot of redundancies. So we solve a lot of the pixel redundancy from VE, but there's more redundancy in long range and long horizon videos. Say, if a character appear in the first clip and then it disappeared, it only reappear at the end of the video, you probably don't need the-- the context, like in the middle of the generation. So you only need that character, where you need. So that's why, I helped build another feature. It's a reference video.Vibhu [00:58:36]: Is it here?Swyx [00:58:36]: is it the same model release or different one?Ethan [00:58:39]: It's a different one.Ethan [00:58:41]: You probably need to search onSwyx [00:58:43]: I'll find itEthan [00:58:43]: X reference to video.Ethan [00:58:46]: So reference video allow you to like upload up to seven images as condition and generate the video. Say, if like I want-- it can, it can be characters or objects or even scenes. Say like I want, I want condition on, Sean's selfie and holding a bladeSwyx [00:59:07]: We have a dogEthan [00:59:08]: or whatever.Swyx [00:59:08]: We put the dog in the thing.Ethan [00:59:09]: you can put them there and the video models will generate the video from and copies the context over. So that can solve a lot of the problems there, like the long context problem. It doesn't need to have a very long context, but it's-- I feel like it's an intermediate solution. The modelSwyx [00:59:29]: It's cheating.Ethan [00:59:30]: the model should be able to like selectively know, where should I draw the references. So say if I want to generate a movie, I generate it autoregressive, like a ten second at a time or something. And now this character appear, I can look back to where it first appear and, bring that back. Yeah, this one, I put the references. Yeah, that's, Optimus, Einstein myself, Annie.Vibhu [01:00:02]: Oddly enough, I used Grok Search to find it, and it pulled your LinkedIn post. But yeah we found it.Ethan [01:00:08]: Interesting.Vibhu [01:00:10]: ButxAI's Underrated Work, Culture, and WatermarkingSwyx [01:00:11]: this is a problem. This is not your fault, but like XAI doesn't communicate all this work that you do very well because they just have the model release and then that's it. But actually, these details are very good.Swyx [01:00:22]: As far as I understand, everything you just described is state-art, like no one else has done it.Vibhu [01:00:30]: A lot of-- yeah, I have a lot moreSwyx [01:00:32]: And then, and then you just put this blog post with the cookies. I'm this is not enough,?Swyx [01:00:37]: but I, obviously this is like the high level numbers that people want to know. But no, okay, soVibhu [01:00:42]: And I wonder, like part of that is also some labs don't share research into what happens. And ifSwyx [01:00:50]: No, but this is literally bragging about how good they are, right?Swyx [01:00:54]: Like, why would you not say that you are capable of extending with full context? this is not a secret sauce. This is like we did the work. yeah, I don't know.Ethan [01:01:02]: different labs have slightly different communication styles.Swyx [01:01:07]: Anyway, if anyone from XAI is listening we are always happy to help you tell your story. Yeah, okay, so you did references, and I think, I think kind of the point you're, you're making is it is sort of like a kludge, right? this is-- you can do seven, but what about 100?Swyx [01:01:23]: Right? Then you need a completely different thing.Ethan [01:01:26]: So I think it's-- this is, a mechanism to, select the context from the history, and you might not put the entire history into the context. for example, there's a paper called Frame Pack, which haveEthan [01:01:41]: a heuristic that the latest history, the last one second, I put the entire history, and the history before that, I would, compress it and makes the video smaller. So they follow this pattern, this build overall pattern that the maximum sequence length is fixed. So the further you are from the current frame, you have a smaller image. So this is just a heuristic. I think it can be more automatic. The model is aware like which history part of it can be select. So this part of the research is actually being actively, worked on by a lot of people. It's also quite interesting. I feel this is actually, this part of long context is a little bit ahead of the LLM part.Ethan [01:02:31]: So for example, like in LLMs, if you-- so contexts keep growing. Let's say if you call tool and the tool call history is extremely long, that's still in context, and keep growing, keep growing. Even if you switch the topic to something else, the whole context was there. There are some agentic harnesses that help you to, say, prune the tool results and, prune Like when you, when you query a file, only show like the top 200 lines or something. Those were very heuristic-driven.Swyx [01:03:08]: For listeners, we did a write-up on the cloud code, leak where there are eight different kinds of pruning, including like you prune the tool results and all that. So you can, you can read up on that kind of thing.Ethan [01:03:17]: I think, one breakthrough in continual learning might be like a way to automatically, manage its own context.Swyx [01:03:27]: These are all heuristics, and they will be replaced by machine learning.Ethan [01:03:30]: InterestinglyVibhu [01:03:32]: TheEthan [01:03:32]: the same thing is being researched in both LLMs and video models.Vibhu [01:03:36]: The interesting thing is also like in the paper you showed, it's actually happening at the model level, right? Compared to like language models, sure, we have base attention, but we'll do our own compression, we'll do our own pruning, which is separate from model error.Vibhu [01:03:49]: Eventually, it all just boils in, hopefully.Swyx [01:03:52]: I think this is a form of like attention, but like also know sort of reasoning attention. I feel like that's different than normal attention.Swyx [01:04:03]: Does that, does that make sense?Ethan [01:04:04]: It's, it's different in the sense that attention, not to mention, set sparse attention aside,

This week we're joined by Steph to talk about fashion and librarianship: how it works, how they intersect, and how fashion is interdisciplinary. We also learn about how you get into this gig! Media mentioned Good general article: Frederiksen, Linda. “Fashion by Design.” Library Journal, vol. 131, no. 10, June 2006, pp. 79–81. EBSCOhost, research.ebsco.com/linkprocessor/plink?id=297645b9-2580-3b4f-bdae-8ee2344a2633. Costume Institute article: Kazmi, Sahar. “A CENTURY OF FASHION: Library Collections Reflect the Evolution of 20th-Century American Style.” Library of Congress Magazine, vol. 13, no. 1, Jan. 2024, pp. 14–19. EBSCOhost, https://www.loc.gov/lcm/pdf/LCM_2024_0102.pdf  ARLIS article: King, Lindsay M., and Russell T. Clement. “Style and Substance: Fashion in Twenty-First-Century Research Libraries.” Art Documentation: Journal of the Art Libraries Society of North America, vol. 31, no. 1, Mar. 2012, pp. 93–107. EBSCOhost, doi.org/10.1086/664912   Library Journal's September By The Numbers focused on fashion libraries: “Fashion and Beauty.” American Libraries, vol. 56, no. 9/10, Sept. 2025, p. 11 https://americanlibrariesmagazine.org/2025/09/02/by-the-numbers-fashion-and-beauty/  All previous transcripts: https://podscripts.co/podcasts/librarypunk  Join the Discord: https://discord.gg/qWPTurTnkT

Olivia Smoliga raced a LCM competition for the first time since the 2024 US Olympic Trials this past weekend in Westmont. Smoliga, 31, raced the 50 back (28.4) and 50 free (24.9), where she made the A-Final in both and placed 3rd in the latter. She also time trialed the 50 fly (26.9). The 2x Olympian gives her perspective on why she decided to return to racing, saying she feels like a whole new person now compared to when she was last competing. Smoliga hasn't been completely away from the pool, however. In the interim, she started In Depth Swim Academy, where she's been able to share what she's learned in and out of the pool with young swimmers across the country.

After the PM's three day visit to China, economist Vicky Pryce discusses his achievements. Reform councillor Jason O'Connell responds to yet another conservative defection to his party, and their plans for the Senedd election. Emeritus Professor Emyr Lewis explains the significance of the assisted dying LCM and why it's wise to wait before the Senedd votes on it. And historian, the Rev. D Ben Rees, talks about his new book about the Welsh of Liverpool.Plaid Cymru's Senedd hopeful, Cllr Mark Hooper and cultural relations specialist Ruth Cocks review the papers.

Today we ask How a Rice Krispie Treat is like a glue gun? and What does LCM stand for? as we taste chewy, crispy congealed goop. We avoid kissing our moms as we learn about cereal sculpting and what the ancients ate before we taste a divisive potato chip. Feel the Fizz of this Snapple Crack! Rice Krispie Treats official recipeSmitten Kitchen Salted Brown Butter Crispy Treats.Molly's Now but Wow - If This Ever Gets Weird doc  Support Spilled Milk Podcast!Molly's SubstackMatthew's Bands: Early to the Airport and Twilight DinersProducer Abby's WebsiteListen to our spinoff show Dire DesiresJoin our reddit Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Ryan Coughenour has had a full journey leading into his senior year of high school. His grandfather was an Oklahoma state champion in the 1940s and went on to swim at Kansas State University (when they had a swim program) as well as compete at the US Olympic Trials in the 1950s. He has supported Ryan through his swimming career and encouraged him to pursue it in college. During the pandemic, when Ryan was in middle school and just starting to swim seriously, he started buying sneakers online and re-selling them for profit. Well, not for profit at first. But eventually, he learned how to make a profit and ended up making quite an income. Coughenour even convinced his coach to let him out in the middle of practice one day in order to secure a pair of valuable shoes, which he ended up re-selling for a profit of $1,100. Bow in his senior year, Ryan has committed to swim at Florida State. He's coming off a summer that saw him rapidly improve in LCM, going best times nearly every time he dove in the pool and capping his season at Summer JRs. Learning to be a leader on his high school and club team, the breaststroke specialist has a lot to look forward to.

O iFood acaba de lançar o Ailo, seu novo assistente de inteligência artificial que permite fazer pedidos diretamente pelo WhatsApp, como se você estivesse conversando com um atendente de verdade. Em entrevista ao Podcast Canaltech, Isabella Piratininga, Diretora de Tecnologia e Inovação do iFood, explicou como o Ailo foi criado, combinando diferentes modelos de IA (como OpenAI e Amazon) com o LCM, large Commerce Model, sistema proprietário do iFood que usa mais de 14 anos de dados para oferecer recomendações personalizadas. Durante o bate-papo, Isabela contou os bastidores do desenvolvimento, os desafios de adaptar a IA ao jeito brasileiro de se comunicar, incluindo sotaques, expressões regionais e até gírias como “sacolé” e “chup-chup” e como a equipe está lidando com as alucinações típicas de modelos generativos. Você também vai conferir: WhatsApp lança IA que resume suas conversas automaticamente, NASA coloca astronautas “dentro de um chip, óiculos da Apple pode rodar dois sistemas ao mesmo tempo, Sam Altman provoca ao falar sobre empregos e IA e Governo distribui parabólicas digitais de graça pelo Brasil. Este podcast foi roteirizado e apresentado por Fernanda Santos e contou com reportagens de André Lourenti, Nathan Vieira, Bruno Bertonzin, Renato Moura, sob coordenação de Anaísa Catucci. A trilha sonora é de Guilherme Zomer, a edição de Jully Cruz e a arte da capa é de Erick Teixeira.See omnystudio.com/listener for privacy information.

In this episode of The Sunday Roast, Phil and Kevin is joined by Charles Archer to unpack a lively week in the markets as sentiment shifts positively going into Q4. The pair discuss the growing “risk-on” appetite, moves in commodities, and the impact of Donald Trump's proposed Middle East peace plan. They also dive into topics like digital IDs, government overreach, and investor protection in the UK markets. The discussion rounds off with the week's movers and shakers, from standout risers such as UK Oil & Gas, Energy Pathways, and Defense Holdings, to sharp fallers like Mast Energy Developments. A packed episode covering geopolitics, small-cap stories, crypto, gold, and everything in between — all wrapped up in classic Roast style. 00:00 - 00:13:20  Weekly News Roundup  00:13:20 #UKOG  00:14:52 #EPP  00:18:42 #BZT  00:23:30 #GPP  00:24:00 #AFP  00:26:10 #ALRT  00:36:15 #LCM  00:36:22 #MAST  00:38:55 #PREM  00:39:54 Gold Market 00:40:58 #COIN  00:44:38 #AAI  00:51:10  #TIR  #ASTR Disclaimer & Declaration of Interest This podcast may contain paid promotions, including but not limited to sponsorships, endorsements, or affiliate partnerships. The information, investment views, and recommendations provided are for general informational purposes only and should not be construed as a solicitation to buy or sell any financial products related to the companies discussed. Any opinions or comments are made to the best of the knowledge and belief of the commentators; however, no responsibility is accepted for actions based on such opinions or comments. The commentators may or may not hold investments in the companies under discussion. Listeners are encouraged to perform their own research and consult with a licensed professional before making any financial decisions based on the content of this podcast. 

Our God is so good to LCM to help us connect with the "ancient paths" that Jeremiah 6 speaks of! With no power, no running water, no video recording... and no complaining! Pastor Spencer & Pastor Wade help set our hearts aright and set our walk on the righteous path of loving God's discipline. God's use of discipline, judgment, and correction are designed to move us beyond our weaknesses & our sin in order that we become the warrior's that God always intended. LCM is rising in tenacity, resolve, and forcefulness that causes us to lean into the discipline & difficulty that are before us. Let this sermon correct your thinking, revive your heart, and set your hand to the work that God has set before you!

Our God is so good to LCM to help us connect with the "ancient paths" that Jeremiah 6 speaks of! With no power, no running water, no video recording... and no complaining! Pastor Spencer & Pastor Wade help set our hearts aright and set our walk on the righteous path of loving God's discipline. God's use of discipline, judgment, and correction are designed to move us beyond our weaknesses & our sin in order that we become the warrior's that God always intended. LCM is rising in tenacity, resolve, and forcefulness that causes us to lean into the discipline & difficulty that are before us. Let this sermon correct your thinking, revive your heart, and set your hand to the work that God has set before you!

This sermon preached by Pastor Eric & Pastor Judah worked to cultivate the attitude of the joyful warrior at LCM. We want to hear from the Lord, well done my good and faithful servant and we know how to do it!

This sermon preached by Pastor Eric & Pastor Judah worked to cultivate the attitude of the joyful warrior at LCM. We want to hear from the Lord, well done my good and faithful servant and we know how to do it!

Dutch backstroker Kira Tousaint recently announced her retirement from competitive swimming. She ended her career as a 3x Olympian, 11x European Champion, and Dutch record-holder in the 50 and 100 back (LCM) as well as the 50, 100, and 200 back (SCM). She was also an NCAA All-American for the University of Tennessee, where she competed from 2015-2017.

On the back of winning the 50 free and 50 fly at the 2025 British Championships and qualifying for her first LCM world championships, SwimSwam sat down with Eva Okaro to discuss her swimming career thus far. Okaro dives into her experience at the 2024 Paris Olympics and the 2024 Short Course World Championships, where she won a silver medal with Britain's mixed 4x100 medley relay. She also talks about her training in Britain and what she's looking forward to about coming to the University of Texas in the fall.

Welcome to Episode 156 of The Scale Model Podcast Sponsored by CultTVMan and Sean's Custom Model Tools HostsStuartTerryGeoffVince HoffmannThanks to our latest Patreon and Buy Me a Coffee Supporters:Check out our What We Like page for lists of what we like.***************************************LATEST NEWS***************************************MAILBAGWe want to hear from you! Let us know if you have any comments or suggestions scalemodelpodcast@gmail.com.***************************************LATEST HOBBY ANNOUNCEMENTSPreview update: Art & plastic of Border Model's 35th scale He-111 H-6.Six releases from Italeri in April1/35th scale M2128 Bradley 105mm mobile gun system from Magic FactoryAirfix 1/72nd Bristol Blenheim Mk.IFSpecial Hobby April NewsletterX-15A-2 in 1/32 scale. IMAM/Romeo Ro.57bis in 1/48 scale and Blackburn Roc Mk.I in 1/72 scaleWhat's new at Scalemates.com**************************************SPONSOR AD #1Cult TV Man***************************************TopicVince Hoffmanhttps://www.vintagemodelplans.com/Build log on facebookhttps://www.rcgroups.com/forums/showthread.php?4696483-1-16-LCM-6-build-loghttps://www.facebook.com/media/set/?set=a.10226022581051799&type=3 ***************************************WHAT'S ON THE BENCHStuart - Nothing at present, it's our club's swap night so I'm doing quite a large stash purge. Already sold a few things.Geoff - drywalling has taken me away from the bench I'm afraid!Terry - Not a lot. Some travel, some organization of the bench. Still doing some work on that Battroid Valkyrie. Mostly addressing seams at this point.***************************************WHAT WE ARE READINGStuart - Not much time this week to read anything.Geoff - same here - so sick of renovations!Terry - still getting through Moonseed and the Apothecary Diaries vol. 2.Vince - Calvin and Hobbs***************************************SPONSOR AD #2Seans Custom Model Tools***************************************THINGS WE'VE SEENArrows on GuardMatters of Scale is working on seamless intakes for the new Airfix Jaguar kit.***************************************THE LAST WORDFor more modelling podcast goodness, check out other modelling podcasts at modelpodcasts.comPlease leave us a positive review if you enjoy what we're doing!Check us out: FaceBook, YouTube, and our very own website. Inbox reviews are available at http://blackfire.ca/We also have merchandise now. Check it out on Redbubble 

Dr. Sam Heidari is the CEO of Lumotive, a groundbreaking semiconductor manufacturer based in Redmond, WA. Sam holds a PhD in electrical engineering from USC and previously served as CEO of Quantenna, where he led the company through a public offering and $1 billion acquisition by ON Semiconductor. At Lumotive, Sam is leading a team that's building award-winning programmable optical semiconductors that improve perception, increase computing power, and enable reliable high-speed communication in various industries. The Light Control Metasurface™ (LCM) is a patented, software-defined photonic beamforming solid-state technology. As the first of its kind, it meets essential needs in various sectors, including 3D sensing and AI computing.See omnystudio.com/listener for privacy information.

Last time we spoke about the Visayas Offensive. In March 1945, the Pacific War raged on. On Iwo Jima, the US Marines, after intense fighting and heavy casualties, declared the island secured. Meanwhile, in northern Luzon, General Clarkson's division advanced towards Baguio, facing fierce Japanese resistance, while General Mullins pushed through Balete Pass. The Japanese army, grappling with severe supply shortages, was forced to evacuate Baguio. In the Visayas, General Eichelberger's forces targeted the Sulu Archipelago and Central Visayan Islands, securing key airfields. The 40th Division landed on Panay, capturing Iloilo, and launched an assault on northern Negros. On Cebu, the Americal Division landed near Talisay, encountering mines but minimal resistance, and secured Cebu City. By April, Allied forces had made strategic advances across the Philippines, overcoming Japanese resistance and establishing crucial airfields. This episode is the Invasion of okinawa Welcome to the Pacific War Podcast Week by Week, I am your dutiful host Craig Watson. But, before we start I want to also remind you this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Perhaps you want to learn more about world war two? Kings and Generals have an assortment of episodes on world war two and much more  so go give them a look over on Youtube. So please subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry for some more history related content, over on my channel, the Pacific War Channel you can find a few videos all the way from the Opium Wars of the 1800's until the end of the Pacific War in 1945.  We have come to the grand final battlefield at last, that of Okinawa. Of course battles are raging in all sorts of other theaters like New Guinea, China, Burma, etc. However as you might imagine its becoming impossible given the week by week format to cover all of this in single episodes. So we are going to hardcore focus on Okinawa for awhile, we will circle back to the other theaters to catch up. Seriously it was the only logical way to do this and honestly in retrospect I wish the entire podcast was campaign by campaign instead of week by week. But I am a mere podcaster following the youtube series of this. But if you want to hear a campaign by campaign series, over at Echoes of War me and my cohost Gaurav are beginning to roll them out. The first series will be the entire Malayan Campaign, and I think after that I might try to do the Philippines. Regardless lets jump into the invasion of Okinawa. As previously noted, the directive issued by the Joint Chiefs of Staff on October 3 designated Okinawa as the final target for invasion, following the establishment of air and naval bases in Luzon and Iwo Jima. Capturing this crucial island would bring the conflict to Japan's doorstep, disrupt the enemy's air communications through the Ryukyu Islands, and flank their maritime routes to the south. Consequently, from these newly established air and naval bases in the Ryukyus, American forces would be able to launch attacks on Japan's main islands and implement a more rigorous sea and air blockade, isolating them from Japanese territories to the south. This made it imperative for the Japanese Empire to maintain control over Okinawa and the Ryukyus. To this end, the 32nd Army, led by Lieutenant-General Watanabe Masao, was formed there by late March 1944. Initially, it comprised four companies and one artillery regiment stationed at Amami Oshima; five companies and one artillery regiment at Nakagusuku Bay; four companies and one artillery regiment at Iriomote Island; along with various garrison units from the 19th Air District in Okinawa.  Under Operation Tei-Go, the Ryukyus and Formosa were to form a long zone of interprotective air bases. These bases were expected to defeat any American sea or air forces sent into the region. To avoid destruction from the air, each base was to consist of a cluster of airfields, such that if one were damaged others could be used immediately. Military and civilian crews were promptly set to work building the numerous fields. 13 base clusters had to be created, stretching in a line from Tachiarai in the northern Ryukyus to Pingting on Formosa in the south. The only remaining tasks for ground forces were the defense of these facilities and their support anchorages and the unenviable work of building the fields. Much of the energy of 32nd Army would be absorbed building these air facilities. This was more difficult since 32nd Army had only two bulldozers and one earth roller. Japan had produced dozers in small numbers at its Komatsu plant since 1943, but few had reached the front. Since soldiers were thus obliged to use shovels, hoes, straw baskets, and horse-drawn wagons, construction was slow. Moreover, because of enemy submarine raiders, it was impossible for the Japanese to deliver the large quantities of fuel, ammunition, and anti-aircraft guns needed to operate the bases. Even more seriously, the planes themselves were not available. Between April and June, the 32nd Army received reinforcements, including the 44th and 45th Independent Mixed Brigades, the 21st Independent Mixed Regiment, and the 27th Independent Anti-Aircraft Artillery Battalion, among other air garrison units. From these forces, Watanabe chose to send the 45th Brigade to establish bases on Miyako Island and Ishigaki Island, while the 21st Regiment was tasked with setting up a base on Tokunoshima. On June 27, the 1st and 2nd Infantry Corps (approx. 4100 men) boarded the Toyama Maru and began the voyage to Okinawa. The 44th Brigade HQ, Artillery and Engineer Units meanwhile boarded other vessels of the Taka-412 convoy. Two days later, while the convoy was sailing east of Tokunoshima, the submarine Sturgeon successfully sank the Toyama Maru with two torpedoes, therefore inflicting the loss of 3724 men and much heavy equipment. Because of this, the 1st Corps had to be deactivated.  Following the fall of Saipan, the Japanese Empire rapidly deployed significant ground forces to the 32nd Army, including the 9th, 24th, 28th, and 62nd Divisions, as well as the 59th, 60th, and 64th Independent Mixed Brigades, and the 27th Tank Regiment. Additionally, the 15th Independent Mixed Regiment was airlifted to bolster the beleaguered 44th Brigade, which was reorganizing its 2nd Corps with local recruits. The 32nd Army Staff wished to use as much of the indigenous population as it could in direct support of the war effort, so on January 1 1945 it ordered total mobilization. All Okinawan males aged 18 to 45 were obliged to enter the Japanese service. 39000 were drafted, of whom 15000 were used as nonuniformed laborers and 24000 as rear-echelon troops called the Home Guard (Boeitai). Many of the Boeitai replaced sea based battalions and rear-area supply units that had been reorganized and equipped for frontline duty. In addition to these, 1500 of the senior boys of the middle schools on Okinawa were organized into Iron and Blood Volunteer Units and assigned to frontline duty. Some of these students had been tried out in the signal service in the autumn of 1944 with good results, so the program was expanded. Since the fall of 1944, 600 senior students of the girls' middle schools also had been given training in the medical service. While most of these troops were sent to strengthen the main defenses at Okinawa, a large portion of the 28th Division was actually assigned to reinforce the garrisons on the Miyako and Yaeyama Islands. The 36th Regiment was dispatched to support the Daito Islands, the entire 45th Brigade was moved to garrison Ishigaki and the rest of the Yaeyamas, the 59th Brigade was tasked with defending Irabu Island, the 60th Brigade was sent to reinforce Miyako Island, and the 64th Brigade was deployed to the Amami Islands. On August 9, Lieutenant-General Ushijima Mitsuru took command of the 32nd Army. His initial strategy was to occupy all of Okinawa with a strong force and eliminate any invading troops at their landing sites. As fierce fighting began at Leyte, intensive training for the troops commenced, including divisional maneuvers at potential American landing points, artillery bombardments of beachheads, and nighttime assaults on bridges. However, on November 13, Tokyo decided to send the elite 9th Division to Formosa to prepare for a movement to the Philippines that ultimately did not occur. The unexpected withdrawal of the 32nd Army's best division disrupted Ushijima's operational plans. Ironically, this situation improved combat efficiency, as it compelled the army to achieve more with fewer resources. Ushijima chose to concentrate most of his forces in the mountainous and easily defensible Shimajiri area, located in the southern part of the island. This strategic location allowed the Japanese to maintain control over Naha port and target the northern airfields with artillery. Similar to the situation in Iwo Jima, this decision marked a departure from the previously favored strategy of a "decisive battle," opting instead for a war of attrition that had proven effective at Peleliu and Iwo Jima. As a result, Ushijima positioned Lieutenant-General Amamiya Tatsumi's reinforced 24th Division at the southern end of the island, Lieutenant-General Hongo Yoshio's reinforced 62nd Division along the central isthmus, Major-General Suzuki Shigeji's reinforced 44th Independent Mixed Brigade on the Hagushi plain, and Colonel Udo Takehiko's Kunigami Detachment, consisting of two battalions from the 2nd Corps, in northern Okinawa. These troop placements were successfully implemented in December; however, concerns arose that the 32nd Army was spread too thin to effectively counter the anticipated enemy invasion. Consequently, on January 15, Ushijima decided to move the 44th Brigade from the Hagushi plain southward to overlap with the 62nd Division's area on the east, significantly shortening the Japanese front. Additionally, one battalion from the Kunigami Detachment was sent to defend Iejima and its crucial airbase, which necessitated the rest of the unit to consolidate its positions and strengthen defenses on Yaedake Mountain in the Motobu Peninsula. Ushijima also had the support of the brigade-sized 5th Artillery Group, led by Lieutenant-General Wada Kosuke; the 21st Field Anti-Aircraft Artillery Corps; the 11th Shipping Group, which included several shipping engineer regiments and sea-raiding battalions; the 19th Air District overseeing various aviation service units; and Rear-Admiral Ota Minoru's Okinawa Naval Base Force, which comprised nearly 9,000 personnel stationed at the Oroku Naval Air Base near Naha. Ushijima had nearly 100,000 troops at his command, with 29,000 assigned to specialized units for anti-aircraft, sea-raiding, and airfield operations. Anticipating the nature of the impending conflict, these well-staffed service units were reorganized for ground combat. The 19th Air District transformed into the 1st Specially Established Regiment, responsible for defending the Yontan and Kadena airfields it had recently constructed and maintained. Most of the service personnel were integrated into the new 1st Specially Established Brigade in the Naha-Yonabaru area, while the sea-raiding base battalions became independent infantry units. Additionally, the remainder of the 11th Shipping Group was restructured into the 2nd Specially Established Brigade on the southwestern part of Okinawa. This reorganization, completed on March 21, bolstered ground combat strength by 14,000 men, leaving only 10,500 of the 67,000 Army personnel in specialized roles. To protect themselves, the Japanese began constructing robust fortifications, tunnels, and cave systems to shield against anticipated enemy bombardments.  Work on the caves was begun with great vigor. "Confidence in victory will be born from strong fortifications" was the soldiers' slogan. The caves meant personal shelter from the fierce bombardments that were sure to come, and they also offered a shimmering hope of victory. The combination was irresistible, and units began to work passionately on their own caves. Enthusiasm was essential because of the great toil it took to create the caves. Just as 32nd Army had only two bulldozers to make airfields, it had no mechanized tunneling equipment at all. Besides lacking cutting equipment, 32nd Army also lacked construction materials. It had no cement, no ironware, and no dynamite. The units had to rely entirely on wooden beams that they obtained themselves to shore up their shafts. This was not necessarily easy because there were no forests in the south of the island where the troops were now stationed. Pine forests were abundant in the mountainous north, however, so each unit was assigned its own lumbering district in the north. Several hundred men from each division were detailed as its lumbering squad. The problem remained, however, of how to move the several million logs that were needed over the 40 or so miles from the forests to the forts. With no railroads and the use of trucks limited by a shortage of gasoline, the solution was for each unit to cut its own logs, then transport them in small native boats called sabenis. The divisions acquired 70 of these, which then plied the waters steadily from north to south. In January 1945, however, the Leyte-based B-24s that began flying over daily for reconnaissance also began strafing the boats. So the waterborne delivery of logs had to be switched from day to night, greatly lowering efficiency. Ushijima aimed to prolong the fight from these positions and decided to destroy the indefensible Yontan and Kadena airfields to prevent their use by the enemy, effectively ending the ineffective air defense strategy initially devised by Tokyo. Meanwhile, Admirals Nimitz and Spruance were preparing for Operation Iceberg, the invasion of Okinawa. Given the expectation that the capture of Iwo Jima and recent air assaults on Japan would concentrate enemy air power around the Empire's core, which would respond aggressively to any attacks on Okinawa, the Americans needed to first neutralize or eliminate enemy air facilities in the Ryukyus, Kyushu, and Formosa to achieve air superiority over their objective. As a result, all available carrier-based and land-based air forces were tasked with this operation, including Admiral Mitscher's Task Force 58, General Kenney's Far East Air Forces, Admiral Hoover's Central Pacific Forward Area, and General Arnold's 20th Air Force. From the 20th Air Force, General LeMay's 21st Bomber Command was assigned to attack Okinawa before moving on to Kyushu and other vulnerable locations in the home islands. Meanwhile, Brigadier-General Roger Ramey's 20th Bomber Command, supported by General Chennault's 14th Air Force, focused on neutralizing Formosa. Aircraft from the Southwest Pacific Area were also set to conduct searches and continuous strikes against Formosa as soon as conditions on Luzon allowed. Additionally, the British carriers of Vice-Admiral Bernard Rawlings' Task Force 57 were tasked with neutralizing air installations on the Sakishima Group in the ten days leading up to the landings. Once air superiority was achieved, Spruance's 5th Fleet was to land Lieutenant-General Simon Buckner's 10th Army, which included Major-General John Hodge's 24th Corps and Major-General Roy Geiger's 3rd Amphibious Corps. The 24th Corps comprised the 7th and 96th Divisions, while the 3rd Amphibious Corps included the 1st and 6th Marine Divisions. As the Commanding General of Expeditionary Troops, Buckner also oversaw the 27th and 77th Divisions and the 2nd Marine Division for special operations and reserve purposes. Furthermore, the 81st Division was held in area reserve. In total, Buckner commanded a force of 183,000 troops, with 116,000 designated for the initial landings. The plan called for Major-General Andrew Bruce's 77th Division to first secure amphibious bases in the Kerama Islands on March 26, with one battalion further securing Keisejima five days later to establish a field artillery group there. On April 1, following a demonstration by the 2nd Marine Division in southern Okinawa, the main landings were set to take place. The 24th Corps and the 3rd Amphibious Corps would land simultaneously on the west coast beaches north and south of Hagushi. After landing, both corps were to quickly advance across the island, capturing the airfields in their designated areas before securing all of southern and central Okinawa. Subsequently, Buckner's forces were to invade and secure Iejima and northern Okinawa. Once this was achieved, there were tentative plans to launch an invasion of Kikaijima with the 1st Marine Division and Miyako Island with the 5th Amphibious Corps. For the invasion, Spruance relied on a Covering Force under his command, along with Admiral Turner's Joint Expeditionary Force. This force included Admiral Blandy's Amphibious Support Force, which comprised minesweepers, UDTs, and escort carriers; Rear-Admiral Morton Deyo's Gunfire and Covering Force, consisting of ten battleships and eight heavy cruisers; Rear-Admiral Lawrence Reifsnider's Northern Attack Force, responsible for landing the 3rd Amphibious Corps; and Rear-Admiral John Hall's Southern Attack Force, tasked with landing the 24th Corps. Similar to Iwo Jima, Okinawa had endured multiple air attacks since October 1944 as part of the preliminary operations for the landings on Leyte, Luzon, and Iwo Jima that we previously discussed. These operations also included air strikes aimed at neutralizing Japanese air power on Formosa. Throughout February and March, Mitscher's carriers and LeMay's B-29s brought the conflict back to the Japanese home islands with a series of strikes that caused significant damage and instilled fear in the population. On March 1, Task Force 58 launched the first strike of the month against the Ryukyus, targeting Amami, Minami, Kume, Tokuno, and Okino, as well as Okinawa. This operation resulted in the destruction of 41 planes, the sinking of eleven vessels, damage to five others, and significant harm to island facilities. In between the main assaults on Japan, the superfortresses frequently targeted key sites in the Ryukyus, leading the beleaguered Japanese forces to refer to these missions as the "regular run." Throughout March, aircraft from the Southwest Pacific and the Marianas conducted nearly daily operations over the Ryukyus and surrounding waters, searching for Japanese shipping and contributing to the isolation of Okinawa by sinking cargo ships, luggers, and other vessels, while American submarines intensified the blockade around the Ryukyus. Meanwhile, after completing extensive training and rehearsals, Task Forces 51 and 55 gathered at Leyte, Task Force 53 assembled in the Guadalcanal-Russells area, and the rest of the 5th Fleet convened at Ulithi. On March 12, Reifsneider's convoy, carrying the 3rd Amphibious Corps, was the first to depart, successfully reaching Ulithi nine days later. On March 18, the tractor group transporting the 77th Division began its journey from Leyte to the Kerama Islands. On the same day, Mitscher's carriers targeted 45 airfields in Kyushu, claiming the destruction of 102 Japanese planes, damaging or destroying 275 on the ground, sinking six vessels, and damaging three more. In response, Admiral Ugaki's 5th Air Fleet launched a counterattack against the carriers Enterprise, Intrepid, and Yorktown, inflicting minor damage while losing 33 aircraft. The next morning, after locating the majority of the Combined Fleet at Kure, Mitscher dispatched 436 aircraft to target naval installations and shore facilities in the Inland Sea. At 06:50 three C6N Saiun “Myrt” recon planes discovered Task Force 58, and by 07:00 Captain Genda Minoru's elite, handpicked 343rd Kokutai had scrambled 63 advanced Kawasaki N1K2-J “George” Shiden-Kai fighters from Shikoku to intercept the Americans. Minutes later, Genda's powerful Shiden-Kais “waded into the Hellcats and Corsairs as if the clock had been turned back to 1942.” Soon the 343rd Kokutai was engaged in a wild maelstrom with 80 US fighters, including VF-17 and VBF-17 Hellcats from Hornet and VMF-112 Corsairs from Bennington. For once the Japanese broke about even, losing 24 fighters and one scout plane to the Americans' 14 fighters and 11 bombers. Nevertheless, Genda's expert but outnumbered 343rd Kokutai proved unable to blunt the American onslaught. Despite facing a formidable intercepting force, they managed to inflict damage on 18 Japanese warships, including the battleships Yamato, Ise, Hyuga, and Haruna, as well as six aircraft carriers. Additionally, one incomplete submarine was destroyed, 97 enemy planes were shot down, and 225 were either destroyed or damaged at Japanese airfields. In response, Ugaki launched a kamikaze counterattack that successfully struck the carriers Wasp and Franklin, causing significant damage and forcing Franklin to head to Pearl Harbor immediately. As Task Force 58 slowly withdrew the afternoon of March 20, a damaged Zero crashed destroyer Halsey Powell, killing 12 and wounding 29. Shortly afterwards friendly anti-aircraft fire started fires aboard Enterprise. At 23:00 eight Japanese torpedo planes unsuccessfully attacked the carriers, while three overnight snoopers were splashed by anti-aircraft fire. Between March 17 and March 20 Ugaki had committed 193 aircraft to battle and lost 161. On March 21 Ugaki dispatched a 48-plane strike, including 16 G4M “Betty” bombers, carrying the very first Ohka (“Cherry Blossom”) suicide missiles. However the combat air patrol (CAP) of 150 Hellcats and Corsairs repulsed them. Franklin, Wasp, and Enterprise, all damaged, steamed to Ulithi as a reorganized Task Group 58.2. Except for April 8–17, when Task Group-58.2 was briefly reestablished, Task Force 58 strength would remain at three Task Groups throughout Iceberg's duration. Ugaki's 5th Air Fleet was meanwhile effectively incapacitated for several weeks, but Ugaki nevertheless reported five carriers, two battleships, and three cruisers sunk, which IGHQ found scarcely credible. Meanwhile, the Mine Flotilla departed Ulithi on March 19, followed two days later by the rest of Task Force 52 and Task Force 54 to support the Kerama operation. On March 23, to prepare for the imminent minesweeping operations of Iceberg, Task Force 58 conducted extensive bombing raids on all known installations in Okinawa, resulting in significant damage and the sinking of 24 vessels and damage to three others over the following five days. Furthermore, Admiral Lee's battleships traversed the cleared area and opened fire on Okinawa, sinking an additional two vessels. With this naval and air support, Blandy's minesweepers and UDTs successfully cleared the route for Rear-Admiral Ingolf Kiland's Western Islands Attack Group by nightfall on March 25. Their primary opposition in the following two days consisted of a series of kamikaze attacks, which caused damage to the battleship Nevada, light cruiser Biloxi, four destroyers, two destroyer minelayers, one minesweeper, and two transports, while the destroyer Halligan was sunk by mines. The next morning, supported by naval gunfire and carrier aircraft, Bruce initiated his first landings. The 3rd Battalion, 305th Regiment landed on the southern beaches of Aka Island, facing sporadic resistance. Meanwhile, the 1st Battalion, 306th Regiment landed unopposed on Geruma Island, which was quickly secured. The 2nd Battalion, 306th Regiment achieved even faster success at Hokaji Island, while the 1st Battalion, 305th Regiment invaded Zamami Island with light resistance. The 2nd Battalion, 307th Regiment encountered minor opposition as it took Yakabi Island. The Fleet Marine Force Amphibious Reconnaissance Battalion scouted Keisejima and found no enemy presence. After encountering some resistance, the units on Aka and Zamami pushed back the enemy garrisons, securing two-thirds of Aka by nightfall and successfully repelling a strong counterattack on Zamami that night. Simultaneously, Deyo's warships and Blandy's carriers began bombarding the demonstration beaches, while minesweepers cleared progressively larger areas around Okinawa, although the minesweeper Skylark was sunk by mines. By March 27, the remaining enemy forces on Aka and Zamami were finally eliminated, and a company took Amuro Island without opposition. Additionally, a company from the 307th moved to Kuba Island, which was quickly secured. At the same time, Bruce continued his main landings, with the 1st and 2nd Battalions of the 306th Regiment landing on the west coast of Tokashiki Island, facing minimal opposition. As the two battalions advanced north along narrow trails toward Tokashiki town, the 3rd Battalion landed to secure the southern part of the island. On March 28, they reached the town, clearing the entire island and concluding the Kerama operation.  In Kerama, "Island Chain between Happiness and Good," the Japanese tradition of self-destruction emerged horribly in the last acts of soldiers and civilians trapped in the hills. Camping for the night of March 28 a mile from the north tip of Tokashiki, troops of the 306th heard explosions and screams of pain in the distance. In the morning they found a small valley littered with more than 150 dead and dying Japanese, most of them civilians. Fathers had systematically throttled each member of their families and then disemboweled themselves with knives or hand grenades. Under one blanket lay a father, two small children, a grandfather, and a grandmother, all strangled by cloth ropes. Soldiers and medics did what they could. The natives, who had been told that the invading "barbarians" would kill and rape, watched with amazement as the Americans provided food and medical care; an old man who had killed his daughter wept in bitter remorse. Only a minority of the Japanese, however, were suicides. Most civilians straggled into American positions, worn and dirty. In all, the 77th took 1,195 civilian and 121 military prisoners. This operation resulted in the deaths of 530 Japanese soldiers, 121 captured, and the neutralization of over 350 suicide boats, with American losses totaling 31 killed and 81 wounded. While this initial operation was underway, the tractor groups of the Southern and Northern Attack Forces left their staging areas in Luzon and Ulithi on March 25, followed by the rest of Spruance's fleet two days later. On March 26 and 27, Rawlings' Task Force 57 conducted a series of strikes on the Sakishima Islands, primarily targeting Miyako. Meanwhile, after a 250-plane raid on the Mitsubishi plant in Nagoya on March 24, LeMay sent 165 B-29s from the 73rd and 314th Bombardment Wings to attack the Kyushu airfields on March 27, facing minimal resistance as they caused significant damage to the Tachiarai Army Airfield, the Oita Naval Airfield, and the Omura aircraft plant. Other bombers from the 313th Bombardment Wing laid aerial mines in the Shimonoseki Strait. The Japanese responded with a raid on Spruance's naval units using aircraft and suicide boats on the night of March 28, resulting in one LCM being destroyed and one cargo ship damaged. On March 29, Mitscher launched another strike against Kyushu, but poor weather conditions led to only minor damage, with 12 vessels sunk and one damaged. By this point, the “largest assault sweep operation ever executed” had cleared the Hagushi beach approaches in 75 sweeps, with minesweepers clearing 3,000 square miles of coastal waters. The following morning, as Task Force 58 once again targeted Okinawa, Deyo's ten battleships and eleven cruisers advanced to bombard Okinawa's defenses and demolish coastal seawalls with increased intensity. At the same time, the 314th sent 12 planes to attack the Mitsubishi engine works in Nagoya overnight. The next day, LeMay dispatched 152 B-29s for his second assault on Kyushu, completely destroying the Tachiarai machine works and heavily damaging the Omura airstrip. On March 31, back in Okinawa, the final underwater demolition operation off the Hagushi beaches was underway while the 420th Field Artillery Group was successfully positioned on Keisejima, prompting a strong reaction from Ushijima's artillery. That morning, a Ki-43 fighter crashed into Admiral Spruance's flagship, the Indianapolis, resulting in the deaths of nine crew members and severely damaging a shaft, which ultimately compelled Spruance to transfer his flag to the battleship New Mexico. Meanwhile, the frogmen completed their last demolition operations at Hagushi, and the final preliminary bombardment of Okinawa and the Sakishima Islands was executed successfully. By the end of the month, over 13,000 large-caliber shells had been fired in the shore bombardment, and approximately 3,095 sorties had been conducted against the Ryukyus. However, effective Japanese concealment prevented significant damage to Ushijima's defenses. As night fell, a vast fleet of transports, cargo ships, landing craft, and warships navigated the final miles of their long journey, successfully meeting off the Hagushi beaches in the East China Sea before dawn on April 1. While Turner's forces prepared for the landing, a fire support group consisting of 10 battleships, 9 cruisers, 23 destroyers, and 177 gunboats began the pre-landing bombardment of the beaches at 05:30, firing a total of 44,825 rounds of shells, 33,000 rockets, and 22,500 mortar shells. In response, the Japanese launched some scattered kamikaze attacks on the convoys, successfully hitting the transport Hinsdale and LST 884. At 07:45, carrier planes from Task Force 58 and Blandy's carriers targeted the beaches and nearby trenches with napalm. Fifteen minutes later, the first wave of amphibious tanks advanced toward the shore at four knots, followed closely by five to seven waves of assault troops in amphibious tractors. Alongside the primary landings, Major-General Thomas Watson's 2nd Marine Division staged a feigned landing on the southeast coast of Okinawa, near Minatoga, aiming to distract the enemy's reserves in that region. Meanwhile, on the main front, supported by rocket fire from LCI gunboats and artillery fire from Keisejima, a nearly continuous line of landing craft advanced toward the beaches at 08:20. Encountering no resistance, the first waves began to land on their designated beaches at 08:30, with additional troops following closely behind. Within an hour, Geiger's 3rd Amphibious Corps had successfully landed the assault elements of the 6th and 1st Marine Divisions north of the Bishi River, while Hodge's 24th Corps disembarked the 7th and 96th Divisions to the south of the river. The lack of significant opposition, coupled with the rapid disintegration of the untrained 5473 airfield service troops of the 1st Specially Established Regiment under heavy air and artillery bombardment, created a sense of foreboding among the men, prompting them to scout the area cautiously. As before, the enemy's primary response consisted of kamikaze attacks on naval units, resulting in damage to the battleships West Virginia and Tennessee, the British carrier Indefatigable, destroyers Prichett and Vammen, the British destroyer Ulster, the destroyer minelayer Adams, and four other vessels. Returning to Okinawa, after ensuring they were not walking into a trap, the troops began advancing inland while tanks and other support units were brought to the beaches. In the north, Major-General Lemuel Shepherd's 6th Marine Division deployed Colonel Merlin Schneider's 22nd Marines on the isolated Green Beaches and Colonel Alan Shapley's 4th Marines on the Red Beaches near Yontan Airfield. The 4th Marines advanced toward Yontan, encountering only scattered resistance, and quickly secured the objective east of the airfield by 13:00. Meanwhile, Schneider's 3rd Battalion moved through Hanza without opposition, but the 2nd Battalion's progress was hindered as it needed to protect its exposed flank, prompting the 22nd Marines to quickly commit its reserve battalion to maintain their momentum. To the south, Major-General Pedro Del Valle's 1st Marine Division landed Colonel Edward Snedeker's 7th Marines on the Blue Beaches and Colonel John Griebel's 5th Marines on the Yellow Beaches just north of the Bishi River. By 09:45, the 7th Marines on the left had advanced through the village of Sobe, their primary objective, while the 5th Marines were positioned 1,000 yards inland. At this point, it was decided to land the reserve battalions of both regiments, along with Colonel Kenneth Chappell's 1st Marines. With forces arranged in depth and reserves positioned to the right and left, Del Valle's units continued to advance steadily over the rolling terrain as the 11th and 15th Marines artillery units were also being landed. At 13:30, the 4th Marines resumed their advance, facing light resistance on the left but becoming overextended on the right while trying to maintain contact with the 7th Marines. As a result, Shapley landed his reserve battalion to fill this gap, while Shepherd also deployed his reserve 1st Battalion, 29th Marines, to secure the critical northern flank, allowing the 22nd Marines to keep advancing eastward. Meanwhile, further south, Major-General Archibald Arnold's 7th Division disembarked Colonel Frank Pachler's 17th Regiment on the Purple Beaches just south of the Bishi River and Colonel John Finn's 32nd Regiment on the Orange Beaches in front of Kadena Airfield. Both regiments quickly ascended the gentle hills at the landing sites and began advancing eastward. By 10:00, the 27th Regiment had patrols at Kadena Airfield, which was discovered to be empty; by 10:30, the front line was crossing the airstrip. Moments later, it advanced 200 yards beyond, heading towards Cholon. Simultaneously, the 32nd Regiment secured the southwestern edge of Kadena and continued along the road to Kozo. To the south, Major-General James Bradley's 96th Division landed Colonel Michael Halloran's 381st Regiment on the White Beaches in front of Sunabe and Colonel Edwin May's 383rd Regiment on the Brown Beaches at the extreme southern flank. Both regiments moved eastward with the same ease as the other units that day, making significant progress towards Momobaru in the north and successfully capturing Chatan in the south. Additionally, all divisional artillery of the 24th Corps landed early; Hodge's reserve regiments and battalions were successfully brought ashore, and by nightfall, direct-support battalions were in position. By the end of the day, over 60,000 men had landed, suffering only 28 dead, 27 missing, and 104 wounded, thereby establishing a beachhead 15,000 yards long and up to 5,000 yards deep in some areas. Shepherd's Marines paused for the night along a line stretching from Irammiya to the division boundary south of Makibaru, which the 1st Marine Division extended further south to Kadena. Meanwhile, the 7th Division advanced nearly three miles inland, destroying several pillboxes but losing three tanks to mines. The 96th Division secured positions along the river south of Chatan, on the elevated ground northwest of Futema, in the outskirts of Momobaru, and in the hills to the northwest and southwest of Shido. Although there were gaps in the lines in several areas, they were filled by reserve units or weaponry before nightfall. Thus, the Battle of Okinawa, seen by most as the final climactic battle of the Pacific War has only just begun. I would like to take this time to remind you all that this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Please go subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry after that, give my personal channel a look over at The Pacific War Channel at Youtube, it would mean a lot to me. As Gandalf the White once said “The board is set, the pieces are moving. We come to it at last, the great battle of our time.” The battle of Okinawa will become the bloodiest campaign America has ever fought. The soul crushing journey has just begun as the Americans end the last stand of the Japanese in the Pacific War.

In his final pre-race check-in, James Sawyer brings his signature low-fuss, high-heart approach to the mic as he prepares to line up for the Delirious West 200-miler. From night runs with his brother to a back-to-back weekend of endurance capped off by a 25K mental health fundraiser walk, James shares how he's gotten his mental game right—even if he still feels like he's “not done enough.”   With a minimalist gear strategy, a no-nonsense approach to planning, and his whole family rallying behind him, James is ready to embrace whatever the trail throws at him. From packing chip sandwiches and Baba's spaghetti to explaining his simple mantra—“just get it done”—this episode is full of charm, honesty, and ultra mindset magic.   Key Highlights: •Weekend warrior: night run + fundraiser walk + early trail session •His brother joins as a night pacer (with bonus responsibility: “don't get us lost”) •Crew strategy with Corey, sleep planning, and a very casual approach to watches & data •Food prep: chips, white rolls, LCM bars & the legendary Baba's spaghetti •Keeping it simple: identical race outfits, minimalist gear, and loads of heart •A reminder that you don't have to be elite to do extraordinary things   Connect with Us: Subscribe for the final pre-race episodes and post-race recaps from the full Delirious crew! Got a race-day ritual or snack combo as unique as chip sandwiches? Drop it in the comments—we want to hear it!   Check out the Delirious WEST event - https://deliriouswest200miler.com.au/ Event Facebook - https://www.facebook.com/groups/1428304207182387   https://fitmindfitbody.co/podcast/

Last time we spoke about the fall of Iwo Jima. General Schmidt pushed through Japanese defenses, facing strongholds like Cushman's Pocket and General Senda's positions, with intense fighting and heavy casualties on both sides. Despite stubborn resistance, the Marines gradually advanced, employing tanks and artillery support. The Japanese, under General Kuribayashi, fought tenaciously, culminating in a final assault on March 26. After brutal combat, Iwo Jima was declared secured, but at great cost: 18,000 Japanese and over 6,800 American lives lost. Meanwhile, in New Britain, Australian forces continued their offensive, capturing strategic positions despite fierce enemy resistance. In the midst of a fierce conflict, Australian battalions advanced into enemy territory, capturing strategic positions while facing heavy resistance. Notable victories included the successful ambushes by the 2/7th Battalion and the capture of key locations like the But airfield. Despite facing fierce counterattacks from the Japanese, particularly at Slater's Knoll, the Australians maintained their momentum. As the Japanese command faced internal crises, the Australians continued their relentless push, ultimately leading to significant territorial gains and weakening enemy forces. This episode is the Visayas Offensive Welcome to the Pacific War Podcast Week by Week, I am your dutiful host Craig Watson. But, before we start I want to also remind you this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Perhaps you want to learn more about world war two? Kings and Generals have an assortment of episodes on world war two and much more  so go give them a look over on Youtube. So please subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry for some more history related content, over on my channel, the Pacific War Channel you can find a few videos all the way from the Opium Wars of the 1800's until the end of the Pacific War in 1945.  Within northern Luzon, by mid-March, General Clarkson's 33rd Division was exploring the western routes to Baguio. Colonel Volckmann's guerrilla unit had taken control of San Fernando and was engaged in combat around Cervantes. Meanwhile, General Mullins' 25th Division had advanced to Putlan, and General Gill's 32nd Division was involved in a fierce battle at Salacsac Pass. Following the capture of Putlan, General Swift instructed Mullins to push through Balete Pass to secure the vital Santa Fe region. On March 12, the 27th and 161st Regiments began advancing north and northwest, successfully clearing the Minuli area and establishing a foothold on Norton Ridge by March 15. At the same time, the 35th Regiment initiated a broad maneuver around Balete Pass via the Old Spanish Road, but this was soon hindered by intense artillery and mortar fire from the positions of the 11th Independent Regiment. The road also required significant engineering efforts to support the outflanking force, leading Mullins to ultimately halt the 35th's assault. However, General Konuma was unaware of this; fearing an outflanking maneuver from Carranglan, he had no choice but to keep around 3,000 of his best troops in this seemingly secure area. On March 15, Mullins ordered the reinforced 161st Regiment to launch a holding attack to the north while the majority of the 27th Regiment executed a flanking maneuver over Myoko Ridge, Mount Myoko, and Lone Tree Hill to encircle Balete Pass from the east. After regrouping at Putlan, the 35th was also tasked with capturing Mount Kabuto to subsequently advance northwest along Balete Ridge and connect with the 27th at Myoko. Mullins launched a renewed offensive on March 16, facing stubborn resistance from the 161st as it gradually advanced toward Norton's Knob, which was successfully captured by the end of the month. On March 22, the 27th began its flanking maneuver, with the 1st Battalion moving east into the elevated terrain south of Kapintalan by March 28, while the 2nd Battalion advanced nearly 3,750 yards up the wooded Myoko Ridge to the east. Meanwhile, the 35th's attempt to encircle the enemy via Kabuto started off well, reaching the northern slope of Balete Ridge on March 22. However, Konuma's recently arrived reinforcements launched a series of strong counterattacks and harassment raids, forcing the 35th to withdraw by March 28. With the other two regiments stretched thin and unable to penetrate the enemy's main defenses, Mullins ordered the 35th to position itself between the 27th and 161st Regiments to assault Highley Ridge from the east. By the end of March, the 161st was advancing northeast along Highley Ridge toward Crump Hill, which finally fell on April 8, bringing the advance to a halt. The 35th supported this effort with an attack on Kapintalan, which was ultimately captured by April 21. Meanwhile, the 27th continued to struggle through the challenging terrain of Myoko Ridge against fierce opposition, not reaching Woody Hill until April 12. This slow progress allowed Konuma to reinforce the Myoko sector with four understrength infantry battalions. Nevertheless, the brave soldiers of the 27th Regiment persevered, securing the Pimple on April 15 and advancing an additional 350 yards northeast by April 21. Looking west, Clarkson aimed to advance battalion combat teams toward Baguio via Route 11, the Galiano road, and the Tuba Trail. However, Swift declined to approve such ambitious plans, limiting the 33rd Division to a more restrained offensive. Although dissatisfied, Clarkson continued the attack along Route 11, managing to reach Camp 3 by the end of March. In the central area, patrols faced no opposition as they approached within a mile of Galiano, but other patrols discovered increasing signs that the Japanese were preparing to defend the Tuba Trail vigorously. Eventually, a robust reconnaissance force secured Bauang on March 19, with patrols then moving east to occupy Naguilan four days later and Burgos by the month's end. Meanwhile, at Salacsac Pass, the 127th Regiment fought from Hill 502 to Hill 504 against fierce resistance, while Gill sent the 2nd Battalion, 128th Regiment up the trail from Valdez toward Imugan, where they were ultimately halted by vigilant Japanese forces. By March 23, the 1st Battalion of the 127th Regiment reached the crest of Hill 504; the 2nd Battalion advanced past it to Hill 505; and the 3rd Battalion managed to position one company at the base of Hill 507D. However, with its forces stretched thin and dwindling, the 127th was unable to capitalize on its seemingly advantageous position, allowing General Iwanaka to launch a strong counterattack on Hill 507D that successfully repelled the 3rd Battalion. As a result, Gill decided to withdraw that unit and the stalled 2nd Battalion, 128th Regiment, while the majority of the 128th relieved the 127th in the Hill 502 sector. Beginning on March 25, the 128th Regiment advanced aggressively eastward, successfully capturing the previously overlooked Hill 503, fully securing Hill 504, and expanding its control over Hill 505 by the month's end. However, on the night of March 31, Iwanaka launched a daring counterattack, reclaiming Hill 504 and nearly taking all the territory east of Hill 502. By April 4, both the 127th and 128th Regiments had sustained significant casualties and could no longer continue the offensive. Consequently, Swift had to order the 33rd Division to relieve the beleaguered 126th Regiment in the Ambayabang and Arboredo River valleys, allowing this unit to reposition north of the Villa Verde Trail to execute a flanking maneuver along the Miliwit River valley. Fortunately for Clarkson, General Krueger persuaded MacArthur to free the 129th Regiment from its duties in Manila, sending it to bolster the now overstretched 33rd Division. Once the rest of the 37th Division arrived at the Baguio front, which Krueger anticipated would happen in early April, Swift could initiate a two-division assault on Baguio. In the meantime, Clarkson promptly dispatched the 129th to Burgos, and by April 1, it had advanced to Salat.  The Japanese opposing the reinforced 33d Division were no longer in the shape they had been at the end of February. The 58th IMB and the 23d Division had both suffered heavy losses during March, losses that probably stemmed largely from lack of food and medical supplies rather than from combat action. By mid-March Japanese supply problems on the Baguio front had progressed from bad through worse to impossible. First, supplies had moved westward over the new Baguio-Aritao supply road far more slowly than anticipated, a development attributable in large measure to Allied Air Forces strikes on that road and along Route 5 north and south of Aritao. Second, operations of the 66th Infantry, along Route 11 north from Baguio, and the activities of the 11th Infantry, , in the Cagayan Valley, had made it virtually impossible for the Japanese to bring any food into the Baguio area from the north. Third, the Japanese tried to do too much with the limited amount of supplies available on the Baguio front. They were attempting to supply 23d Division and 58th IMB troops along the MLR; send certain military supplies north up Route 11 for the 19th Division; feed 14th Area Army headquarters and a large civilian population in Baguio; and establish supply dumps north and east of the city against the time of eventual withdrawal. Almost inevitably the principal sufferers were the front-line troops. By mid-March the best-fed Japanese combat troops on the Baguio front were getting less than half a pound of rice per day as opposed to a minimum daily requirement of nearly two and a half pounds. Before the end of the month the troops on the MLR were down to less than a quarter of a pound of rice a day. Starvation and diet-associated diseases filled hospitals and sapped the strength of the combat units. Generally, effective frontline strength was far lower than reported ration strength indicated. Medical supplies were consumed rapidly, and by the end of March, for example, there was virtually no malaria phophylaxis left in Baguio area hospitals. Looking upon the situation on the Baguio front with frank pessimism, Yamashita in mid-March directed inspection of terrain north, northeast, and east of the city with a view toward preparing a new defense line. His attitude became even plainer when, on or about 30 March, he ordered Japanese civilians and the Filipino puppet government to evacuate Baguio. Indeed, the future on the Baguio front was so bleak by the end of March that almost any other army would have withdrawn to new defenses forthwith, thereby saving troops for future battle. But not so the Japanese. Yamashita decided that the existing MLR would be held until the situation became hopeless. At the end of March that portion of the MLR held by the 23d Division was still intact, and the 58th IMB was busy deploying additional strength along its section of the line. One independent infantry battalion was on high ground north of Route 9 at Sablan; and another held defenses at Sablan. A reinforced company was at Burgos and, less that company, another independent infantry battalion held reserve positions at Calot, a mile and a half southeast of Sablan. One understrength battalion was responsible for defending the rough terrain from Sablan six miles south to Mt. Apni, where a tie-in was made with the right flank of the 23d Division. Maj. Gen. Bunzo Sato, commanding the 58th IMB, expected that the emphasis of any Allied drive in his sector would come along Route 9, but he did not neglect the other approach in his area, the Galiano road. Since the understrength battalion stationed astride the road was not strong enough to withstand a concerted attack, he directed his main reserve force, the 1st Battalion of the 75th Infantry, 19th Division, to move west out of Baguio to defenses at Asin. This step left in Baguio a reserve force of roughly three provisional infantry "battalions," which together probably could not muster over 750 effectives. In the Salacsac area, as casualties in the western pass were rapidly increasing and Allied air and artillery strikes made it nearly impossible for the 2nd Tank Division to transport supplies, Iwanaka concluded that his forward positions were nearly untenable. He began planning to redeploy forces to defend the eastern pass. Consequently, when the 126th Regiment launched its new offensive on April 5, it faced unexpectedly light resistance, quickly capturing Hills 518 and 519 within two days and cutting off a Japanese supply route leading north from Hill 504 across the eastern slopes of Hill 519 and up Mount Imugan. However, Iwanaka swiftly recognized the emerging threats and sent reinforcements to intercept the 126th's advance, delaying the capture of Hills 511 and 512 until March 13 and effectively halting further progress. At the same time, the reorganized 128th Regiment resumed its assault eastward on April 7, successfully retaking the Hill 504-505 sector and capturing Hill 506 by April 10. The following week, the 128th fought tenaciously to secure this elevated ground against fierce resistance, managing to push all the way to Hills 506B, 507C, and 507D with their remaining strength, nearly securing the western pass by April 17. Meanwhile, after the fall of San Fernando, General Krueger instructed Volckmann to advance inland along Route 4 toward Bontoc. However, the Provisional Battalion established at Cervantes could only withstand the relentless enemy pressure until April 4, when it was ultimately forced to retreat into the hills northwest of the town. Consequently, Volckmann had to quickly deploy the 121st Regiment to barrio Butac in preparation for an eastward push along Route 4. After a week of intense back-and-forth combat, this elite guerrilla unit succeeded in establishing footholds along the northern parts of Lamagan and Yubo Ridges. Over the next few days, the 121st made slow and arduous progress, ultimately gaining control of Route 4 nearly to the southeastern edge of Bessang Pass by mid-April. Further south, by mid-March, General Wing's 43rd Division had effectively secured the Antipolo sector, while General Hurdis' 6th Division was advancing well toward Mount Baytangan. At this point, General Hall's 11th Corps assumed responsibility for operations against the Shimbu Group, now rebranded as the 41st Army. However, he quickly decided to maintain the offensive against General Yokoyama's left flank, with the 20th and 1st Regiments focusing on an eastward push alongside the 43rd Division. Meanwhile, General Noguchi had completed his withdrawal to the Sugarloaf Hill-Mount Tanauan line, where he would receive reinforcements from elements of the Kogure Detachment. Simultaneously, General Kobayashi was struggling to stabilize his left flank, bolstered by one reserve battalion. On the morning of March 15, the American offensive resumed, with the 103rd Regiment continuing its assaults on Benchmark 7 Hill to secure Route 60-A, while the 172nd Regiment launched its initial attacks toward Sugarloaf Hill, facing fierce resistance. On March 17, the 1st Regiment renewed its advance toward Baytangan, initially making good progress and digging in about a mile west-southwest of the mountain's summit. However, during the night, a barrage of mortar fire followed by an infantry counterattack forced the Americans to retreat in disarray. As a result of this setback, the 1st Division began to advance eastward more cautiously, facing determined resistance and heavy mortar fire. By March 22, it was only slightly closer to Baytangan's crest than it had been five days earlier. The 20th Regiment on the left also encountered strong resistance, managing to reach a point a mile and a half west of Baytangan by March 22. At the same time, a company maneuvering to the north established a foothold on a wooded ridge overlooking the Bosoboso Valley. Meanwhile the 103d Infantry finally overran the defenses on Benchmark 7 on 18 March, killing about 250 Japanese in the process. The Japanese battalion there had delayed the 103d's attack toward Mt. Tanauan until the morning of 18 March, but by evening of that day the regiment's troops had begun swarming up the bare, rocky, southern and southwestern slopes of the mountain. Over the next three days, American forces slowly advanced through a complex of caves and bunkers until they secured the mountain's summit, while other units captured Benchmark 23 Hill and patrolled northward into the southeastern part of the Bosoboso Valley. By March 19, the 172nd Division was halted, containing Sugarloaf to the west. They bypassed it to the north and east, initiating new assaults toward Mounts Yabang and Caymayuman, making significant progress to the east but less so to the north by March 22. Despite concerns that Hall's four exhausted regiments might lack the strength to turn the 41st Army's left flank, the ongoing pressure and the failure of previous counterattacks ultimately compelled Yokoyama to order his threatened units to withdraw to new positions east of the Bosoboso River. Therefore, when the 6th and 43rd Divisions resumed their attacks on the morning of March 23, the withdrawal was already in full progress. Over the next three days, the 1st and 20th Regiments faced only scattered and disorganized resistance, allowing them to establish positions over a mile north and south of Baytangan along the ridgeline that overlooks the Bosoboso Valley. Meanwhile, the 172nd Regiment successfully captured Mount Yabang and most of Mount Caymayuman, while the 103rd Regiment advanced quickly northward, taking barrio New Bosoboso, Mount Balidbiran, and Benchmark 21 Hill. On March 27, the 1st Regiment secured the crest of Baytangan, and the 172nd Regiment eliminated the last organized opposition at Sugarloaf Hill. With Yokoyama's left flank collapsing, Hall could focus on capturing Wawa Dam and destroying the remaining elements of the Kobayashi Force west of the Bosoboso River, a mission assigned to the 6th Division. During this initial offensive, approximately 7,000 Japanese soldiers were killed since February 20, while American casualties included 435 killed and 1,425 wounded. Hurdis' initial strategy for capturing Wawa Dam involved the 1st and 20th Regiments advancing northward to clear Woodpecker Ridge and Mount Mataba, while the 63rd Regiment conducted diversionary attacks on the western slopes of Mataba. This new offensive began on March 28 but quickly faced intense small arms, machine-gun, and mortar fire, leading to a back-and-forth struggle against fierce enemy resistance. By April 3, the 20th Regiment had advanced less than half a mile toward Mataba, and the 1st Regiment had gained only 250 yards to the north. General Hurdis had hoped his attack, directed against the Kobayashi Force southern flank, would be far more successful, but the Kobayashi Force, rapidly and efficiently, had reoriented its defenses, which it had laid out primarily to face an attack from the west. The force's two remaining provisional infantry regiments, the Central and Right Sector Units, were still relatively intact, and the Central Sector Unit, bearing the brunt of the 6th Division's offensive, had recently been reinforced by remnants of the Left Sector Unit and elements of the Shimbu Group Reserve. Other factors bearing on the 6th Division's slow progress were the declining strength and deteriorating combat efficiency of its infantry regiments. The 20th Infantry could muster only 2,085 effectives on 3 April; some of its rifle companies were reduced to the combat strength of platoons. The situation within the 1st Infantry, with an effective strength of 2,150, was little better. As of 3 April the commanders of both regiments rated their units' combat efficiency only as "fair," the lowest ranking of three terms each had employed since the Lingayen Gulf assault.To bolster protection for Hurdis' right flank, the 103rd Regiment also captured Hill 1200 on the east bank of the Bosoboso River by the end of the month. Due to the slow progress of Hurdis' offensive, on April 5, Hall assigned General Cunningham's Baldy Force, which consisted of the 112th Cavalry Regiment and the recently arrived 169th Regiment, to take control of the area north of Mount Oro. This move would free up most of the 63rd Regiment to support the 20th Regiment and continue the assault north toward Mataba, while the 1st Regiment maintained its position along Woodpecker Ridge. From April 6 to 9, the 63rd made only limited progress to the east; however, on April 10, it shifted its focus to an offensive on the western slopes of Mataba, quickly securing the southwestern quarter of the mountain with minimal resistance. In response, Kobayashi promptly redirected his forces back to Mataba from the north-south ridge to prevent the 63rd from reaching the mountain's summit until April 17. At the same time, the 1st Regiment resumed its advance along Woodpecker Ridge but again faced strong enemy opposition, resulting in only limited gains. Looking further south, by March 23, the 187th Glider Regiment and the 511th Parachute Regiment had advanced to Santo Tomas and Tanauan but were unable to completely clear these areas. Meanwhile, the 158th Regiment secured Balayan, Batangas, and the Calumpan Peninsula, pushing toward Mount Macolod, where they were ultimately halted by significant forces from the Fuji Force. At this stage, as Krueger planned to launch an offensive into the Bicol Peninsula using the 158th, he needed to relieve the 11th Airborne Division with the barely rested 1st Cavalry Division. This division quickly took control of the Santo Tomas-Tanauan area, while General Swing's units maneuvered around the west side of Lake Taal to relieve the 158th at Macolod. As a result, General Griswold renewed his offensive on March 24. The 187th attempted to attack Macolod but was unsuccessful. A task force composed of units from Swing's other two regiments advanced quickly north toward Lipa, only to be halted at the hills southeast of the town. The 8th Cavalry captured Santo Tomas after a fierce battle and took Tanauan two days later. The 7th Cavalry advanced about five miles east into the corridor between Mounts Maquiling and Malepunyo, while the 12th Cavalry pushed along Route 21, moving about four miles beyond Los Baños. On March 27, Swing's task force finally overcame the enemy defenses southeast of Lipa, but it was the 8th Cavalry that ultimately secured this strategic town two days later. Griswold's successful offensive also forced around 2,000 troops from the Fuji Force to retreat along Route 21 and through the Santa Maria Valley to join Yokoyama's forces in the mountains east of Manila. In response, Krueger devised a plan to prevent the potential influx of Japanese reinforcements around the eastern and northern shores of Laguna de Bay. He instructed Hall to clear the northern shore of the lake, block the Santa Maria Valley, and secure Route 21. While the 187th continued its assault on Macolod, Griswold ordered his other units to push eastward to secure Laguna de Bay and Tayabas Bay. Accordingly, on March 30, the 103rd Regiment advanced in small increments along the northern shore of Laguna de Bay, reaching Siniloan by April 4. The 12th Cavalry moved to Calauan and then south along a secondary road toward San Pablo, encountering strong enemy positions that wouldn't be overcome until April 5. The 5th and 7th Cavalry Regiments fought through the Maquiling-Malepunyo corridor against determined but disorganized Japanese resistance, successfully occupying San Pablo by April 2. Additionally, elements of the 188th Glider Regiment pushed east through the mountain corridors, reaching Tiaong on April 3 and Lucena three days later. By April 6, the 5th Cavalry and the 103rd Regiment had established contact at Pagsanjan, effectively isolating the 41st Army. The next day, patrols from the 11th Airborne Division headed north from Lucena, while 1st Cavalry Division patrols departed from Pagsanjan heading south. They successfully linked up at Lucban by April 10 and then advanced eastward to Mauban. At the same time, a company from the 188th traveled along Route 1 across the Bondoc Isthmus, reaching Atimonan on April 11. Griswold's patrols also moved towards Mount Malepunyo, where the Fuji Force was preparing for its final stand. By April 16, preliminary assaults had concentrated enemy resistance around Mount Mataasna-Bundoc. Meanwhile, after a week of intense aerial bombardment, Brigadier-General Hanford MacNider's 158th Regiment successfully landed at Legaspi Port with minimal opposition on April 1. They quickly secured the port and the nearby airfield, then advanced to Daraga before moving south along Route 1 to occupy the Sorsogon Peninsula, where they encountered machine-gun fire from the Mount Bariway-Busay Ridge. The following morning, the troops had to retreat east of Daraga, and in the coming days, the 158th faced tough fighting in the challenging, jungle terrain to overcome resistance in the Daraga area. Concerned about delays in occupying the Sorsogon Peninsula, MacNider landed his anti-tank company at Bacon on April 6, which occupied Sorsogon without opposition. Meanwhile, the 2nd Battalion continued its overland advance, reaching Bulan by April 12, where they targeted a significant Japanese concentration. Simultaneously, MacNider's other two battalions attacked Camalig on April 11 and began their unsuccessful assault on enemy positions in the Cituinan Hills. By April 1, the 188th had successfully eliminated the last organized resistance in the rugged hills south of Ternate. Meanwhile, the 2nd Battalion of the 151st Regiment launched an assault on Caballo Island on March 27 but could not fully dismantle the entrenched enemy positions until April 13. The Japanese in the pits and tunnels created an almost insoluble problem for the 2d Battalion, 151st Infantry. The Japanese had so emplaced their weapons, which included machine guns and mortars, that they controlled all approaches to the mortar pits but could not be reached by American artillery or mortar fire. When the 151st Infantry concentrated its mortar fire against the pits' entrances, the Japanese simply withdrew into the tunnels. When the American fire ceased--at the last possible moment before an infantry assault--the Japanese rushed out of the tunnels to man their weapons. Tanks were of no help to the American troops. From positions near the rim of the pits the tanks were unable to depress their guns sufficiently to do much damage to the Japanese. If the tanks tried to approach from above, they started sliding down Hill 2's slopes into the pits. No combination of tank, artillery, and infantry action proved of any avail, and the 151st Infantry had to give up its attempts to take the Japanese positions by assault. On 31 March engineers tried to pour diesel oil into one of the tunnels connecting the mortar pits, employing for this purpose a single ventilator shaft that was accessible to the 151st Infantry. Nothing came of the effort since it was impossible to get enough oil up the steep slopes of the hill to create a conflagration of significant proportions within the tunnels. Nevertheless, burning the Japanese out seemed to promise the only method of attack that would not risk the unduly heavy casualties of a direct infantry assault. No one, of course, wanted to throw away the lives of experienced troops on such an insignificant objective. Finally, the commander of the 113th Engineers, 38th Division, suggested pumping oil up the hill from the beach through a pipeline from a ship or landing craft anchored at the shore line. The Allied Naval Forces happily fell in with this idea and supplied the 151st Infantry with two oil-filled ponton cubes; the Allied Air Forces provided a 110-horsepower pump and necessary lengths of pipeline and flexible hosing; and the 592d Engineer Boat and Shore Regiment came through with an LCM to carry the pump and the ponton cubes. On 5 April over 2,500 gallons of diesel fuel were pumped into the pits and tunnels through the ventilator and were then ignited by white phosphorus mortar shells. "Results," the 38th Division reported, "were most gratifying." A huge flash fire ensued, followed by a general conflagration and several explosions. The engineers repeated the process on 6 and 7 April, and on the latter day carefully lowered two large demolition charges through the ventilator shaft and placed another at an accessible tunnel entrance. Set off simultaneously, the three charges caused an enormous volume of flames and several terrific explosions. For the next few days the 2d Battalion, 151st Infantry, tried to persuade a few Japanese who had lived through the holocausts to surrender and also executed a few infantry probing attacks. On 13 April a patrol entered the pits and tunnels, killed the lone surviving Japanese, and reported the positions cleared and secured. Following this, El Fraile was targeted, with Company F of the 151st and the 113th Engineers effectively neutralizing Fort Drum using gasoline and explosives. On April 16, troops from the 1st Battalion of the 151st Regiment attacked Carabao Island, facing no opposition. Looking towards the Visayas, after capturing Palawan and Zamboanga, General Eichelberger set his sights on the Sulu Archipelago, where new airfields were to be established. On March 16, a reinforced company from the 162nd Regiment landed unopposed on Basilan Island and spent the next two days searching Basilan and nearby islets, finding no signs of Japanese forces. On April 2, the reinforced 2nd Battalion of the 163rd Regiment landed on Sanga Sanga Island, successfully clearing the Tawi Tawi Group by April 6. The remainder of the regiment made an unopposed landing near Jolo Town on April 9. Over the next two days, the Americans drove approximately 2,400 men of the 55th Independent Mixed Brigade from the heights immediately south and southeast of the town to secure a nearby airstrip. However, the Japanese retreated to more fortified hill masses further inland, where they had long prepared their defenses and began to resist fiercely. After initial attacks by Colonel Alejandro Suarez's guerrillas failed to breach the Japanese positions, the 1st Battalion of the 163rd Regiment joined the fight and managed to overrun the enemy defenses by April 22. On April 25, the 3rd Battalion launched an assault on Lieutenant-General Suzuki Tetsuzo's final positions at Mount Tumatangus, which were ultimately subdued by May 2. However, the remaining Japanese forces continued to engage in guerrilla warfare. By this time, airfields in Zamboanga and Sanga Sanga had become operational to support the forthcoming invasion of Borneo.The first field at Zamboanga was a dry-weather strip 5000 feet long, completed on March 15 and immediately put to use by Marine Corps planes. The field, named Calarian Drome, could not answer the need for an all-weather strip 6000 feet long. Accordingly, engineers constructed a new strip, which the Marine aviators based there called Moret Field, about a mile to the east, and had it ready for all-weather operations by May 16. Used primarily by Marine Air Groups 12, 24, and 32, Moret Field was also employed by a 13th Air Force night-fighter squadron, an emergency rescue squadron, and 13th Air Force B-24s and P-38s staging through for strikes against Borneo. Marine Corps planes on March 16 executed the first support mission flown from a field in the Zamboanga area, covering the landing on Basilan Island. Later, Marine Corps planes from Zamboanga flew support for the Tawi Tawi and Jolo operations and undertook pre-assault bombardment and cover for the invasion of eastern Mindanao. While 13th Air Force planes executed most of the support for the invasion of Borneo, Marine Corps B-25s from Zamboanga also flew some missions. At Sanga Sanga Island there was a Japanese coral-surfaced strip about 2800 feet long. Engineers repaired and extended this strip to a length of 5000 feet by May 2, when fighters of the 13th Air Force began moving to Sanga Sanga from Palawan to provide close support for the initial landings on Borneo. These US Army planes were replaced in mid-May by units of the Royal Australian Air Force, which employed the all-weather Sanga Sanga field during later operations on Borneo. Finally, a Japanese field 3800 feet long on Jolo Island was repaired and used for aerial supply and evacuation operations in support of ground troops throughout the Sulu Archipelago. Eichelberger's next target was the Central Visayan Islands. For the Panay-Guimaras-northern Negros operation, codenamed Victor I, he assigned General Brush's 40th Division, excluding the 108th Regiment, which had recently been assigned to Leyte. The 40th Division departed Lingayen Gulf on March 15 aboard ships from Admiral Struble's Task Group 78.3. After a brief stop at Mindoro, they arrived at Panay before dawn on March 18. Following a short bombardment by destroyers, the 1st and 3rd Battalions of the 185th Regiment landed unopposed about twelve miles west of Iloilo, where they were joined by Colonel Macario Peralta's guerrillas, who already controlled much of the island. The 185th quickly expanded its beachhead against light, scattered resistance and began advancing along the coastal road toward Iloilo in the afternoon, forcing the 170th Independent Battalion to retreat to the mountainous interior. With Iloilo secured by March 20, G Company of the 185th Regiment successfully captured Inampulugan Island on March 22, while other elements of the regiment cleared the unoccupied Guimaras Island by March 23. Additionally, Brush decided to assign the 2nd Battalion of the 160th Regiment to garrison Panay alongside Peralta's guerrillas. No pursuit of the 170th Independent Battalion was made, allowing the Japanese to continue their guerrilla activities until the end of the war. Meanwhile, for the operation against northern Negros, Brush planned to launch an assault with the 185th Regiment on March 29, followed by the majority of the 160th Regiment the next day, while keeping the 503rd Parachute Regiment in reserve for potential airdrops. Opposing him, Lieutenant-General Kono Takeshi's 77th Brigade had a total of 15,000 troops but was prepared to retreat into the mountains of north-central Negros for a prolonged defense, leaving only token forces in the coastal plain to delay American advances. On March 29, the 185th Regiment landed unopposed near Pulupandan and quickly secured a bridge over the Bago River. The 185th then spread north and east, with the 160th Regiment following, successfully securing nearly the entire coastal plain of northwestern Negros by noon on April 2. As they closed in on Kono's inner fortress, the Americans overran the main Japanese outposts while the 511th was landed to bolster the assault. On April 9, Brush launched his general offensive, with his three regiments advancing slowly into rugged terrain where the Japanese held significant defensive advantages.  On June 4 General Kono, realizing that his remaining forces were incapable of further sustained effort, directed a general withdrawal deep into the mountains behind his broken defensive lines. The surviving Japanese dispersed into small groups seeking food and hideouts and trying to avoid contact with Colonel Abcede's guerrillas who, under the direction of the 503rd Parachute Regiment, took over responsibility for the pursuit of Kono's men. On June 9 the 503rd then relieved all elements of the 40th Division in northern Negros. By that date the Japanese had lost over 4000 men killed. Kono lost another 3350 troops, mainly from starvation and disease, before the end of the war. After the general surrender in August 1945, over 6150 Japanese came down from the mountains to turn themselves in, joining about 350 others who had been captured earlier. In all, about 7100 Japanese lost their lives in northern Negros, pinning down the equivalent of an American infantry division for over two months. The 40th Division's casualties for the operation, including those of the attached 503rd Parachute, totaled approximately 370 men killed and 1035 wounded. Meanwhile, Eichelberger's final objective in the Central Visayas was Cebu, assigned to Major-General William Arnold's Americal Division. For Operation Victor II, Captain Albert Sprague's Task Group 78.2 was set to land the bulk of the division at Cebu City while Colonel James Cushing's guerrillas secured the water sources. Opposing them were Rear-Admiral Harada Kaku's 33rd Naval Special Base Force and Major-General Manjome Takeo's 78th Brigade, which together comprised approximately 14,500 troops, including the 173rd Independent Battalion and the 36th Naval Guard Unit stationed at Cebu City. Additionally, around 750 soldiers from General Kataoka's 1st Division were positioned in northern Cebu. Although the command situation on the island was chaotic, General Suzuki's 35th Army headquarters in Leyte had been evacuated by two large motorized landing barges between March 17 and 25. Taking control of all forces on Cebu, Suzuki appointed Manjome as the de jure commander in the Cebu City area while leaving Kataoka in charge of northern Cebu, as he prepared to retreat to Mindanao. General Manjome designed his defenses so as to control--not hold--the coastal plains around Cebu City, and for this purpose set up defenses in depth north and northwest of the city. A forward line, constituting an outpost line of resistance, stretched across the first rising ground behind the city hills 2.5 to 4 miles inland. A stronger and shorter second line, the main line of resistance, lay about a mile farther inland and generally 350 feet higher into the hills. Back of this MLR were Manjome's last-stand defenses, centering in rough, broken hills 5 miles or so north of the city. Anticipating that American forces would attempt to mount wide envelopments of his defensive lines, Manjome set up one flank protective strongpoint in rugged, bare hills about 3.5 miles north of barrio Talisay, on the coast about 6 miles southwest of Cebu City, to block the valley of the Mananga River, a natural axis of advance for forces enveloping from the south and west. Similarly, he established strongpoints on his left to block the valley of the Butuanoan River, roughly 4 miles northeast of Cebu City. Against the eventuality that the American invading forces might land north of Cebu City and strike into the Butuanoan Valley, Manjome set up another flank protective position in low hills overlooking the beach at Liloan, 10 miles northeast of Cebu City. Manjome did not intend to hold the beaches, but at both Talisay and Liloan, the best landing points in the Cebu City region, he thoroughly mined all logical landing areas. The Japanese also constructed tank barriers along the shore line and planted tank traps and minefields along all roads leading inland and toward Cebu City. The inner defense lines were a system of mutually supporting machine-gun positions in caves, pillboxes, and bunkers. Many of these positions had been completed for months and had acquired natural camouflage. Manjome's troops had an ample supply of machine guns and machine cannon and, like the Japanese on Negros, employed remounted aircraft and anti-aircraft weapons. Manjome had some light and heavy mortars, but only a few pieces of light artillery. For the rest, however, Manjome's forces were far better supplied than Kono's troops in northern Negros. After an uneventful journey, Task Group 78.2 and the Americal Division arrived off Cebu in the early hours of March 26. Following an hour of bombardment by three light cruisers and six destroyers from Admiral Berkey's Task Force 74, the leading waves of the 132nd and 182nd Regiments landed unopposed on beaches just north of Talisay at 08:30. However, the landing was chaotic, as Japanese mines just a few yards beyond the surf line disabled ten of the first fifteen LVTs. Fortunately for the Americans, Manjome had chosen to withdraw from the beaches to establish inland defenses, resulting in minimal casualties. Once they cleared the beach minefields by 10:00, Arnold's leading units cautiously advanced through abandoned defenses toward the main highway to Cebu City, ultimately stopping for the night about a mile and a half south of their objective. The following day, the infantry secured Cebu City without opposition and on March 28 proceeded to clear Lahug Airfield and Hill 30 to the north. The Americans began their assault on Go Chan Hill on March 29, during which Company A of the 182nd Regiment was completely annihilated by the explosion of an ammunition dump located in caves along the hill's eastern spur. Fueled by a desire for revenge, nearly the entire 182nd returned to the attack on March 30 and successfully captured Go Chan Hill. Meanwhile, the 132nd Regiment cleared the coastal plains area north to the Butuanoan River, further securing the city's water supply sources by April 2. Unopposed, troops from the 132nd also successfully landed on Mactan Island, quickly securing an airstrip. In the meantime, as guerrillas had already taken control of much of Masbate, the 2nd Battalion of the 108th Regiment successfully landed on the island by April 7. Facing minimal opposition, the Americans pursued the scattered Japanese remnants through the hills and jungles of Masbate, killing approximately 120 Japanese soldiers by May 4.   I would like to take this time to remind you all that this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Please go subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry after that, give my personal channel a look over at The Pacific War Channel at Youtube, it would mean a lot to me. On March 29th, the Americans attacked Go Chan Hill, suffering heavy losses. Seeking revenge, they captured it the next day, securing water supplies and defeating Japanese forces in Masbate. Then in April, Allied forces advanced strategically, overcoming fierce Japanese resistance, securing key positions, and establishing airfields, culminating in significant victories across the Philippines.

This week on the SwimSwam Breakdown, we discuss the women's NCAA Championships and LCM swimming from around the globe.

This week on the SwimSwam Breakdown, we discuss the Stanford women leaving early from the ACC Championships and the plethora of LCM racing around the world we saw this weekend.

LCM is One House. A house that has been drawn into oneness by the Spirit and Word of God. His spirit and his word has unified us to him and to the body of Christ. We are therefore called to do the same for all around us,

LCM is One House. A house that has been drawn into oneness by the Spirit and Word of God. His spirit and his word has unified us to him and to the body of Christ. We are therefore called to do the same for all around us,

303 Episode #471 Best of 2024 Welcome to Episode #471 of the 303 Endurance Podcast. We're your hosts Coaches Rich Soares and April Spilde. Thanks for joining us for another week of endurance news, coaching tips and discussion. Shoutouts to: @genucan @ironmantri @303triathlon @tridottraining @tridottrainingsystem #ironmantri #cycling #triathlon #swimbikerun #Iamtridot #tridotambassador #tridotcoach   Show Sponsor: UCAN UCAN created LIVSTEADY as an alternative to sugar based nutrition products. LIVSTEADY was purposefully designed to work with your body, delivering long-lasting energy you can feel. Whether UCAN Energy Powders, Bars or Gels, LIVSTEADY's unique time-release profile allows your body to access energy consistently throughout the day, unlocking your natural ability to finish stronger and recover more quickly!    In Today's Show Announcements Ask A Coach Get Gritty Ironman Confessions Triathlon News/Updates - X-Mas Gift Exchange     Announcements:   Upcoming Programming in January - In January we are going to bulletproof you and your season with a focus on strength and mobility. In fact we have Erin Carson on board to share her expertise on the podcast.    FulGaz Virtual Group Ride Schedule Dec. 21 - IM Kona 8am MT The FulGaz Experience Lessons Learned and Adjustments Dec. 28 - IM Cozumel 8am MT Jan. 4 - IM Nice 8am MT   Ask A Coach Sponsor: G2G Coaching Triathletes, are you ready to level up and see what's possible with smarter training? At Grit2Greatness Coaching, we're here to set you up for success with TriDot's game-changing platform.   When you click Coach April's or Coach Rich's TriDot Sign-Up link, you'll get a white-glove onboarding experience designed to help you hit the ground running. We'll walk you through every step of the process—from setting up your TriDot account to understanding how the platform tailors your workouts to fit your goals and lifestyle.  Your onboarding will include 2-weeks of free access and then you can decide on plans as low as $14.99 per month.  Get started with us because we've been where you are and know how to bridge the gap between where you're starting and where you want to be. With TriDot and our support, you'll train smarter, gain confidence, and crush your goals.   Ready to get gritty? Start your journey today—because every epic finish line starts with the first step!   Ask A Coach:  Coach Rich, as you know, I had my CSS test this morning and it was my first in a 25 meter (LCM) pool. I maintained my swim dot score, and definitely gave it my all. The only issue is that I know I left a lot of meat on the bone because of my flip turn technique, especially during the 400 meter portion. In 2025, my swim goal is to focus on improving my flip turns so that I am faster doing them then turning at the wall. What tips do you have to help me master my flip turns? Train With Coach Rich: Coach Rich Soares Rich.soares@tridot.com Rich Soares Coaching TriDot Signup - https://app.tridot.com/onboard/sign-up/richsoares RunDot Signup - https://app.rundot.com/onboard/sign-up/richsoares   Train with Coach April: Coach April Spilde April.spilde@tridot.com Grit2Greatness Endurance Podcast Podcast Series - Apple Podcasts TriDot Signup - https://app.tridot.com/onboard/sign-up/aprilspilde   Best of 2025:   It has been a 303 tradition to have our final episode of the year to reflect on the year. We are going to do this in two parts:   Part 1: Our Favorite “Tri Related” Things    Rich's: Pro Race at any distance: Paris Olympic MTR Personal Race or Event: Colorado's Ride Personal Training Day: Pike Peak Book: The One Thing by Gary Keller and Jay Papasan Podcast Episode: Gwen Jorgensen Races Local and Mark Allen Reset Tips   April's Pro Race at any distance: Definitely the XTERRA North American Championships at Oak Mountain State Park, Pelham AL–getting to see my faves all in one place and race on the same day! Personal Race or Event: Boulder Sunset Tri - Athena 2nd place finish and flew out of the water looking like swamp thang, LOL Personal Training Day: Riding the Falcon Trail on the Air Force Academy Book: The Courage to Be Disliked by Ichiro Kishimi and Fumitake Koga (Highly recommend you listen to this one). A self-help book that encourages readers to embrace their individuality and live authentically, regardless of others' opinions, by applying the philosophical concepts of Alfred Adler, which emphasize the idea that true happiness comes from accepting oneself and not seeking validation from others, essentially finding courage to be disliked if it means staying true to yourself; the book is structured as a dialogue between a young man and a philosopher discussing these ideas, allowing readers to explore the concepts through conversation. Podcast Episode: Ep. #271 Building Powerful Triathlon Habits & Episode #461 Preventing/Rehabbing Injury 9 Year End Reflection Questions     Get Gritty Tip: Welcome to this week's “Get Gritty” tip!   This is where we dive deep into the nuts and bolts of mental toughness, resilience, and building a winning mindset—not just for peak performance in triathlon but for life itself. After all, the best athletes are those who inspire others and lead by example. This week, we're drawing inspiration from Richard Barrett's Seven Levels of Consciousness as outlined in his book, “A New Psychology of Human Well-Being.”  Every decision you make—from pushing through the last mile of a race to handling a tough conversation with a teammate—stems from your current awareness. Recognizing that and learning how to elevate yourself is a game changer. These levels offer a fascinating way to think about how we grow mentally and emotionally, and how our consciousness evolves to guide our actions. Let's break it down:   The Seven Levels of Consciousness map your mental and emotional state across different dimensions. The first three levels—Survival, Relationship, and Self-Esteem—are where fear-based, limiting beliefs often hold us back. Imagine you're mid-race, and your energy plummets. If your mindset shifts to Survival mode, you may feel overwhelmed and anxious, questioning your ability to finish. Sound familiar? These thoughts are rooted in fear, not possibility. Here's the gritty part: when you recognize these fear-based triggers, you gain the power to shift your mindset. For example, instead of spiraling into doubt when you're struggling on the bike or in open water, ground yourself by focusing on what's in your control—your breathing, cadence, or a mental mantra like 'calm and strong.' This technique helps you climb from Survival consciousness to higher levels like Transformation or Internal Cohesion, where you're aligned with your purpose and values, not paralyzed by fear. For coaches, this model is equally transformative. When you understand where your athletes are mentally, you can tailor your guidance to meet them where they are. An athlete stuck in Survival consciousness after a DNF might need reassurance and small, actionable steps to rebuild their confidence. By recognizing these patterns, you help them develop resilience and guide them toward realizing their potential. Ready to go deeper? Level 1: Survival Consciousness   *Triathlete Perspective*:   At this level, you're focused on meeting your most basic needs—hydration, nutrition, rest, and overall health. Think about the first time you trained for a triathlon. Were you worried about how far you could swim without panicking or whether you'd have the energy to finish a workout? Survival consciousness is about building a foundation of physical security. For new triathletes, it's essential to master these basics without succumbing to fear. Trust your training plan and learn to prepare for challenges like staying fueled on a long ride or handling a flat tire on race day. *Coach's Perspective*:   A coach operating at this level helps their athletes build confidence in the fundamentals—proper gear, safety in open water, and pacing basics. It's about creating an environment where athletes feel safe to grow without being overwhelmed by survival fears.  Level 2: Relationship Consciousness   *Triathlete Perspective*:   Once your survival needs are met, you start seeking connection—belonging to a training group or feeling supported by friends and family. At this stage, having training partners or a triathlon club is critical. It's where you begin to feel like you're part of a team and gain strength from the relationships you've built. *Coach's Perspective*:   A great coach fosters a sense of community. They emphasize the importance of collaboration, such as relay races or group workouts, where athletes feel supported. This level is also where trust is built between coach and athlete, helping athletes feel safe to share fears or insecurities. Level 3: Self-Esteem Consciousness   *Triathlete Perspective*:   This level is about proving to yourself that you belong in the sport. Maybe you're chasing your first podium finish, hitting a PR, or finally calling yourself a "triathlete" without hesitation. However, the flip side of this level is battling self-doubt. A bad workout or race can trigger those nagging thoughts of not being good enough.  *Coach's Perspective*:   Coaches at this level focus on building confidence. They set realistic goals for their athletes to achieve small wins, reinforcing a sense of accomplishment. A coach might celebrate a new PR or remind athletes that even showing up is a victory on tough days. Level 4: Transformation Consciousness   *Triathlete Perspective*:   At this stage, you start asking more profound questions. Why am I doing this? What does triathlon mean to me? This is where athletes often discover their *why*—whether it's to push boundaries, inspire others, or simply find joy in the process. Transformation consciousness is about stripping away ego-driven motives and aligning your training with core values. *Coach's Perspective*:   Coaches help guide athletes toward this level by encouraging reflection. This might involve post-race debriefs, journaling about their journey, or exploring how triathlon fits into their broader life goals. Level 5: Internal Cohesion Consciousness   *Triathlete Perspective*:   Triathletes find meaning and fulfillment in their training and races at this level. It's not just about crossing the finish line but about how the sport contributes to your well-being. You start to notice how the discipline, focus, and resilience from triathlon spill over into other areas of your life.   *Coach's Perspective*:   Coaches help athletes connect their training to a larger purpose. This might mean encouraging them to mentor newer triathletes, participate in charity events, or reflect on how triathlon enhances their relationships and careers. Level 6: Making a Difference Consciousness  *Triathlete Perspective*:   You're not just in the sport for yourself at this stage—you want to give back. Maybe you're mentoring a beginner, volunteering at a race, or advocating for inclusivity in triathlon. Your focus shifts to making a positive impact on others.   *Coach's Perspective*:   Coaches at this level inspire athletes to think beyond personal goals and contribute to the community. They might lead initiatives to promote youth triathlon programs or encourage athletes to support causes that align with their values. Level 7: Service Consciousness   *Triathlete Perspective*:   This is the pinnacle of consciousness, where triathlon becomes a way of life and an expression of your deepest self. Training and racing are no longer "work" but a form of play, joy, and connection. You're fully aligned with your purpose, using the sport as a platform to live authentically and inspire others.   *Coach's Perspective*:   A coach at this level is no longer just teaching athletes—they're transforming lives. They model selfless service and show that triathlon can be a force for good, whether it's inspiring resilience, building community, or fostering personal transformation. Here's the gritty takeaway: every level builds on the last. To grow as athletes and humans, we need to move beyond basic survival or simple performance metrics and tap into reflection, connection, and purpose. When you level up your consciousness, you level up your life—inside and outside the sport.   Triathlon is the perfect metaphor for life—challenging, rewarding, and constantly evolving. By learning to identify and elevate your level of consciousness, you don't just get stronger physically; you build the mental toughness that helps you overcome any obstacle. Remember, growth begins with awareness.  So, this week, ask yourself: what level of consciousness are you operating from right now? How can you elevate it? Whether you're a triathlete or a coach, climbing these levels could be the key to unlocking your potential.   That's it for our “Get Gritty” tip of the week! Stay tuned for more practical strategies to build resilience and mental toughness in every episode.   Today's Fun Segment: Ironman Confessions Description: A no-holds-barred look at the funny, embarrassing, or downright weird moments athletes have experienced during training or races. Think: porta-potty stories, open-water freak-outs, or “bonking” in the most awkward situations. Example: “I ate someone else's gel at an aid station…on purpose.” Example: “I have sucked the sweat out of the bill of my own visor.” Truth or Lie? Example: “Drank out of a dog bowl at the end of a driveway” Truth or Lie? We want to hear from you! Share your Ironman Confessions with us at ****– and we'll share them on the show!  

Faces of Lions delivered by Pastor Wade and Pastor Matthew, speaks to the core of who LCM is. We can take great confidence that as we approach our Father, and meet with him face to face, it causes us to go out with empowerment to do His very will. Like King David, the one thing that we desire is to be with our King and to be in this very presence, walking through the elements of His Tabernacle, until we meet with Him in the most holy place. In His presence is where we are changed, and from what we have experienced in His presence, we then take that to the rest of the world full of power and confidence.

After the glorious sendoff of Elder Boz and Natalie to the nation of Romania. We have taken some time to reflect on what Adonai has done here at LCM and through the One Association in the last nine months. What extraordinary times and achievements we are living in. Through mutual sacrifice, a tenacity to fight, and an all out pursuit of Kingdom, we are experiencing and living in times that we only used to dream about. With all that said, “We're Just Getting Started”. It is time to go deeper and fan into flame what we have been entrusted. We're Just Getting Started is the attitude toward our mission to win the world for Jesus Christ. We are more accountable today than we have ever been. Therefore, we will not shrink back, back up, or be seduced away from accomplishing all that King Jesus said that we will. If we can look back and see what has been accomplished in the last nine months, then how much more the next nine months! We are excited for the days that are ahead, because We’re Just Getting Started.

A Father’s Gift delivered by the pastors of LCM, was a reminder of how precious the gift of Elder Boz and Natalie are to the body of LCM, and the One Association as a whole. In their final sermon with us, we celebrated their life and the exemplary model they have been to us during their time here at LCM. We have been blessed by their lives over the years and are honored to have sent them off to Romania, to complete the work that Adonai has ordained for them to do in the Balkans. As we heard from Elder Boz in his final words to the congregation of LCM, we were reminded of the serious call to discipleship, sacrifice, and complete abandonment of all other pursuits to complete the call on our lives. What Elder Boz and Natalie have demonstrated for in their faith, is of the same quality, caliber, and quarry of the faith of Abraham and Sarah. Although we will miss Elder Boz and Natalie, the greatest honor we can show towards them, is to live in the same way they have modeled for us, with sacrificial actions that strengthens the next generation, so that they may further the work of the Kingdom.

In this week's episode, the first of Season 6, Greg and Patrick visit with Dr. Ethan McCormick, an Assistant Professor of Educational Statistics and Data Science in the School of Education at the University of Delaware. After catching up on Ethan's international adventures they discuss his recent work on using growth factors as predictors of distal outcomes and how pretty much everything he expected to find came out exactly the opposite. Along the way they also discuss chain sawing family memories, the 31st 1st day of school, Irish goodbyes, barn doors, ridiculous footnotes, blatant plagiarism, done and dusted, throwing R code at your advisor, landing a triple axel, umlauts, being proudly uneducated, hiding in the bathroom, and in fairness to us.Stay in contact with Quantitude! Web page: quantitudepod.org Twitter: @quantitudepod YouTube: youtube.com/@quantitudepod Merch: https://www.redbubble.com/people/quantitudepod/shop

Jeff Kessler, FACHE Allergy Choices This week I sit down with Jeff Kessler to discuss allergies and sublingual immune therapy. Jeff Kessler is a business leader and a thought leader in the allergy therapy space. His relevant backstory goes back to 1999 when he had the fortunate opportunity to work on a project to select and implement an Electronic Medical Record system for the Allergy Associates of La Crosse (AAOL) clinic in La Crosse WI. Fortuitously for his daughter who at the time suffered profoundly with allergies, asthma and eczema, this opportunity presented a new course of therapy that would eventually be taken for the entire family leading to allergic disease modification. Enthralled with the therapy and company, Jeff dedicated his career in the healthcare industry from that point on to the allergy treatment approach that he witnessed change thousands of lives including his families. The growing AAOL organization became linked to Allergychoices, an education company dedicated to teaching healthcare providers around the U.S. on the protocol (La Crosse Method - LCM) that had been perfected since 1970. Jeff has served as President of Allergychoices since its formation and its services have grown to help providers deliver immunotherapy using the LCM to their patients directly at the primary care level. Bringing together AAOL, the most advanced allergy immunotherapy clinic in the U.S. and Allergychoices, capable of delivering the benefits of disease modifying allergy immunotherapy to anyone, anywhere, the distinctive competencies of these companies continue to help hundreds of thousands of allergy sufferers of all types, including foods. Jeff earned a bachelors in science from the University of Wisconsin LaCrosse before he received his MBA in Strategic Planning from University of Minnesota's Carlson School of Management in 1987. Please enjoy my conversation with Jeff Kessler, Dr. M

When we, LCM, are on mission, no snake bite can stop us. Like Paul, we can shake off snakes or sin, and keep moving towards the goal of the mission. We do this by remembering who we are as sons of God & not letting go of the things God has spoken to us. LCM is a church full of men and women who are on mission, doing the last thing the Lord told them until he speaks the next. This is how we will advance our kings agenda in our lifetime.

LCM bars are awful (unless they're a sponsor). We rank World Cup Soccer Balls. Listen Live on the Nova Player App Follow us on Instagram - TikTok - FacebookSee omnystudio.com/listener for privacy information.

We are at war, and we are determined to win! “Fire Fight” is a call to action and to reignite the fire of God inside of us that we may see the fire of God inside of the one life in front of us. Through the lives of men like Moses, Jacob, and Elijah, we see that men of God are not made in a vacuum. But rather have had an experience with the living God of consecrated devotion, that results in an outward focus to see the transformation of the next life. We are LCM. We are at war, and we will see the nations won for the great and glorious name of Yeshua!

We are at war, and we are determined to win! “Fire Fight” is a call to action and to reignite the fire of God inside of us that we may see the fire of God inside of the one life in front of us. Through the lives of men like Moses, Jacob, and Elijah, we see that men of God are not made in a vacuum. But rather have had an experience with the living God of consecrated devotion, that results in an outward focus to see the transformation of the next life. We are LCM. We are at war, and we will see the nations won for the great and glorious name of Yeshua!

We have been asking ourselves a question, how did we get here? How did we become a part of this family of believers here at LCM? It was because we were at one point, the one life that was in front of an empowered saint of God who took the great commission seriously. This call to make disciples and to step out boldly to be a witness of the gospel, is what has built LCM. It is the ancient path that we are walking on. What has been done for us, we will freely do for others. We will take the one life in front of us and give them all that we have. Until we see the transformative power of the living God at work inside of them. This is “The One Life in Front of You”.

We have been asking ourselves a question, how did we get here? How did we become a part of this family of believers here at LCM? It was because we were at one point, the one life that was in front of an empowered saint of God who took the great commission seriously. This call to make disciples and to step out boldly to be a witness of the gospel, is what has built LCM. It is the ancient path that we are walking on. What has been done for us, we will freely do for others. We will take the one life in front of us and give them all that we have. Until we see the transformative power of the living God at work inside of them. This is “The One Life in Front of You”.

“1 More Life, 1 More Family, 1 More Nation” is a call to action for LCM. We have the heavily mandate to win the world for Jesus Christ. We know who we are, and we are to make known to other sons what they have been given the right to. The die for us has been cast and we have stepped over the line. Our decision has been made, and we are on the offensive for King Jesus. We will go out and find 1 more life. We will go out and find 1 more family. And it will result in 1 more nation taken for the gospel. We will give all until we drop, preach until all know, and proclaim until all are empowered by His spirit. This is LCM and we are not ashamed to fight for the gospel of Christ.

“1 More Life, 1 More Family, 1 More Nation” is a call to action for LCM. We have the heavily mandate to win the world for Jesus Christ. We know who we are, and we are to make known to other sons what they have been given the right to. The die for us has been cast and we have stepped over the line. Our decision has been made, and we are on the offensive for King Jesus. We will go out and find 1 more life. We will go out and find 1 more family. And it will result in 1 more nation taken for the gospel. We will give all until we drop, preach until all know, and proclaim until all are empowered by His spirit. This is LCM and we are not ashamed to fight for the gospel of Christ.

Last time we spoke about the fall of Mogaung. General Vinegar Joe's strategy against Myitkyina had turned into a brutal struggle. Stillwells forces were besieging the town while the Chindits battled for Mogaung, diverting Japanese reinforcements in the process. Despite heavy losses, the Chindits captured Naungkyaiktaw with the help of flamethrowers. Gradually Mogaung fell, but the credit was wrongly attributed to the Chinese while it should have gone to the Chindits. Mytikyina withstood endless attacks and at Biak the Americans repelled numerous Japanese assaults. The 2nd battalion, 162nd regiment tossed over to help, but this further complicated supply issues on the plateau. Progress was deeply hampered by water shortages, difficult jungle terrain and supply roads needing repairs. Lastly the battle for Lone Tree Hill saw absolute mayhem and horror, something the Americans and Japanese would never forget. This episode is the Invasion of Saipan Welcome to the Pacific War Podcast Week by Week, I am your dutiful host Craig Watson. But, before we start I want to also remind you this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Perhaps you want to learn more about world war two? Kings and Generals have an assortment of episodes on world war two and much more  so go give them a look over on Youtube. So please subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry for some more history related content, over on my channel, the Pacific War Channel you can find a few videos all the way from the Opium Wars of the 1800's until the end of the Pacific War in 1945.    The use of US fleet and escort carriers during the landings on Hollandia and Aitape had convinced Admiral Toyoda's staff that the American naval force could be drawn into decisive naval battle off of New Guinea. On June 1st, however, a reconnaissance flight had discovered a major build up of enemy vessels at Majuro, suggesting that a new invasion was just days away. The Imperial General Staff reeled with shock in response to this, not knowing whether the Allied objective was Biak, Palau or Saipan. Aboard Toyoda's flagship Oyodo in Tokyo Bay, the naval staff debated, and the debate degenerated into argument. The truth was that no one had a “feel” for the American strategy except one lone staff intelligence officer, Commander Nakajima Chikataka. He said it had to be Saipan. But his fellow staff members scoffed at this. Even the Americans did not have such power that they could mount an operation the size of the three-pronged New Guinea strike as just a feint, they argued. They  further argued that an enemy airfield at Biak would allow the Allies to bomb the Philippines and Palau with their long-range B-24 Liberators. And the American carriers were in evidence off New Guinea, unlike at Majuro. Thinking that Nakajima must be seeing things, Toyoda thus decided that Biak had to be the place. The Japanese unleashed Operation KON, diverting an enormous sum of their naval and air resources to bolster Biak in response to the perceived allied advance.   However, in actuality, Admiral Spruance was gathering the largest Pacific force yet seen at Pearl Harbor, comprising approximately 535 ships, ranging from battleships to tugs, carrying a total of 127,500 personnel of Lt-General Holland Smith's 5th Amphibious Corps.  They were going to hit Saipan, Tinian and Guam. Designated “Operation Forager” it was to be the end game of Admiral Nimitz Central thurst into the Pacific. Taking these would see strategic airfields only 1400 miles away from Tokyo in the hands of the allies. From there the US Army Air Force could begin their long wished bombing campaign against the Japanese home islands. For the Japanese, Saipan in particular was considered a “home island”. Japan had seized Saipan during WW1 from the German Colonial Empire and she heavily invested in “Japanizing” the island. The Japanese government had encouraged southern Japanese and Okinawans to colonize Saipan, introducing Japanese language and culture to the indigenous peoples there. It was more than just another Imperial asset, in many ways Saipan had become a home island. Part of the new operation would include Admiral Mitscher commanding the Fast Carrier Task Force 58 stationed at the Marshalls, along with Vice-Admiral Charles Lockwood overseeing 19 submarines and Rear-Admiral Ralph Christie leading 9 submarines operating across the Pacific and Southeast Asia. Operation Forager entailed Admiral Turner's Task Force capturing Saipan and Tinian sequentially with the 2nd and 4th Marine Divisions, followed by the 3rd Marine Division and the 1st Provisional Marine Brigade attacking Guam. The 27th Division was reserved for Saipan and the 77th Division for Guam. The outcome on Saipan would determine subsequent operations, with the Southern Landing Forces also held in reserve for the Northern Landing Forces until June 25. Lieutenant-General Holland Smith's strategy for the Saipan invasion involved landing four assault regiments simultaneously on four beaches in the Charan Kanoa area. Following three days of continuous air and naval bombardments, Major-General Thomas Watson's 2nd Marine Division's 6th and 8th Marine Regiments were to land north of Afetna Point on the Red and Green Beaches, while Major-General Harry Schmidt's 4th Marine Division's 23rd and 25th Marine Regiments would land south of Afetna Point on the Blue and Yellow Beaches. Thomas Watson was then going to seize the Charan Kanoa Airstrip, afterwards he would sing northeast towards Mounts Tipo Pale and Tapotchau while Schmidt secured Charan Kanoa village, the Aslito Airfield and the islands southern end before moving north to take part in clearing the remainder of Saipan. Alongside this, Smith planned to dispatch the reserve 2nd and 24th Marine Regiments of Lt Colonel Walter Stuart and Lt Colonel Guy Tannyhill over to Scarlet and Black Beaches in the north to divert enemy attention. Two innovations would be tried for Forager. The 4th Marine Division had recently and optimistically accepted the proposal to employ amphibian tanks inland. This was a large part due to the unit's experiences on low and level Roi-Namur, during the Kwajalein assault. The 2nd Marine Division on the other hand, having fought on rugged Guadalcanal and experienced the limitations of LVTs during the Tarawa assault, were more pessimistic. General Watson convinced General Smith that he would only allow the amphibian tanks to advance inland far enough to clear the immediate beach area and that only the first troop-carrying LVT wave would follow them to discharge their troops beyond the beach. All subsequent waves would discharge troops on the beach and not proceed beyond the Tractor Control Line. The reason for this pessimism was that the Marines expected the terrain to be rougher than some predicted. They feared control would be lost over troops embarked in LVTs, concentrated groups of men would be exposed to fire, and that more amtracs would be lost to enemy fire inland. They were needed to return to the Transfer Control Line on the reef's lip where support troops boated aboard LCVPs would be picked up and run to shore in the amtracs. Furthermore, LVTs are not designed for cross-country movement. They have comparatively narrow tracks and low ground clearance, making it easy for them to "belly-out" on rough ground, rocks, and stumps. Amphibian tanks are poor substitutes for medium tanks for the same reasons, and in addition they have very thin armor and a high profile. The Marine divisions commenced training for Saipan in March, conducting a significant rehearsal between May 14 and 20 off Maui. With all arrangements in place, Admiral Turner's invasion fleet was poised to depart Hawaii by late May. The slower LSTs were scheduled to set sail for Eniwetok on May 25, followed by the troop transports departing Hawaii five days later. Turner's vessels were slated to gather at Eniwetok on June 6, coinciding with the departure of Admiral Mitscher's carriers from Majuro for the Marianas. Between June 8 and 10, Turner then embarked from Eniwetok in echelons, experiencing a relatively uneventful voyage towards Saipan. Meanwhile, the Japanese expedited the deployment of 31st Army units to the Central Pacific, successfully positioning most of their line units at designated locations by the end of May. Additionally, certain disparate units of the 31st Army underwent reorganization into seven independent mixed brigades and five independent mixed regiments to enhance chain of command and troop cohesion.   This, however, did not allow adequate time to prepare inland defenses in multiple lines, especially in the case of the Marianas. The larger elevated islands within the Marianas offered strategic advantages for establishing multi-layered defenses, creating strongholds on key terrain features, and positioning reserves centrally for potential counterattacks. However, the Japanese were unable to fully exploit these opportunities due to various factors. Insufficient troop deployments occurred until less than two months prior to the US assault, leaving limited time for comprehensive inland defense preparations. Many coastal defense and anti-aircraft guns were discovered still packed and alongside incomplete positions, due to a shortage of construction materials for fortified installations. Consequently, the high ground at the island's center remained unprepared for defense, and several beach defenses were unfinished. There was an overestimation of the natural ruggedness of the terrain to impede American advances. Despite this, some coastal defense guns were strategically placed within earlier reinforced concrete structures. While Saipan and Tinian lacked significant artificial underwater or beach obstacles, mines were sporadically utilized on their shores. The Japanese also suffered from a lack of anti-aircraft ammunition to counter the enemy air superiority. Over on Saipan, Lieutenant-General Saito Yoshitsugu commanded a significant military presence including the 43rd Division, the 47th Independent Mixed Brigade, the 9th Tank Regiment, the 1st Battalion of the 18th Regiment, and various support units totaling 25,500 personnel. Despite its apparent strength, a convoy carrying reinforcements for the 43rd Division was ambushed by two American submarines between June 4 and 6, resulting in the loss of five transports and nearly decimating the 118th Regiment. Saito, overseeing Saipan as commander of the Northern Marianas Army Group, was joined by Admiral Nagumo's Central Pacific Area Fleet headquarters, bolstering defense with an additional 6,200 naval personnel. Additionally, General Obata, stationed at Saipan's army headquarters, delegated authority to his chief of staff, Major-General Igeta Keiji, as he inspected defenses in Palau during the American landings. Saito organized Saipan's defense into five sectors, anticipating landings on the island's west and northwest coasts, Tanapag Harbor, and the north shore of Magicienne Bay. The Japanese strategy relied on tank-infantry counterattacks launched at night from designated positions to repel enemy landings. However, American bombardment of Saipan commenced before Saito's forces could fully fortify their positions. Large numbers of heavy weapons, ranging from 200mm mortars to 5-inch coastal defense guns and literally dozens of searchlights and antiaircraft guns still remained in naval depots or loaded on railroad cars or still packed in cosmoline near partially excavated gun positions. The high ground in the island's center had received little attention in the defensive preparations; and, indeed, beach defenses themselves were in many cases far from complete. In so far as training in defensive warfare is concerned, Japanese forces displayed a marked deficiency; there was little evidence of organization of the ground, the principal airfield (Aslito) was virtually undefended against ground troops, and their counterattack plans were poorly conceived. On the afternoon of June 11, Mitscher's Task Force 58, began a relentless three and a half day heavy bombardment against Saipan, Tinian, Guam, Rota, and Pagan. This onslaught, featuring a 225-plane fighter sweep, caught the Japanese off guard in the Marianas and resulted in the destruction of 150 enemy aircraft, securing air superiority for the rest of the campaign. On June 12th and 13th, 216 carrier bombers hit Guam, Saipan, and Tinian, while Admiral Hoover's B-24s started pounding the southern Marianas airfields. You might be raising your eyebrows, the B-24's belonged to Task Force 57, under Vice-Admiral John Hoover, which was currently formed by the 5th Air Force over on Cape Gloucester and New Guinea; the 7th Air Force and 4th Marine Aircraft Wing on the Marshall and Gilbert Islands; and the 13th Air Force on the Admiralty Islands. Concurrently, two Japanese convoys were intercepted, leading to the sinking of 16 enemy vessels. The first of these, was a formation of 20 ships, fleeing on a northerly course 125 miles west of Pagan Island. They were attacked on the afternoon of the 12th by planes of Rear Admiral William K. Harrill's Task Group 58.4. Further  south, 135 miles west of Guam, the other convoy of six ships became the target for a special attack mission sent by Rear Admiral Joseph J. Clark, commanding Task Group 58.1. Between the two attacks 12 cargo ships, three submarine chasers, and one patrol-torpedo boat were sunk. Additionally, on June 13th, Mitscher's fast battleships shelled Saipan, though with limited success. The following day, Turner's fire support ships joined Task Force 58 in bombarding Saipan's beaches and installations, effectively neutralizing most of the Japanese coastal defenses and anti-aircraft positions. Despite some damage inflicted on the destroyer Braine and battleship California, Saito's defenses were not completely destroyed, indicating that the preliminary bombardment was not entirely effective. Meanwhile, underwater demolition teams surveyed the landing beaches, clearing them of mines and obstacles and creating gaps in the reefs for the amphibious approach. Smith's reserves staged a demonstration off Tanapag to divert Japanese attention, while carrier groups were dispatched to prevent enemy aircraft staging through the Bonins, successfully attacking Iwo Jima, Chichi-jima, and Haha-jima. To deter the Japanese from utilizing the Bonins in the north as a staging area for aircraft, two task groups led by Clark's 58.1 and Harrill's 58.4 set sail from Saipan on June 14th for an offensive. This marked the furthest advance into Empire waters by a carrier striking force at that time. Task Groups 58.1 and 58.4 refueled on June 14th and were then dispatched that night toward Chichi and Iwo Jima. Airstrikes on Japanese airfields in those locations commenced the following afternoon despite challenging weather conditions. Despite facing strong resistance from the Japanese, American fighter sweeps and bombing runs resulted in the destruction of 28 out of 37 Zeros engaged, with an additional seven destroyed on the ground, while only two American aircraft were lost. Adverse weather hampered morning strikes on the 16th, but the afternoon sortie of 54 American aircraft encountered Japanese planes on the runway, claiming 63 destroyed, though this wasn't confirmed by Japanese sources. Only two American aircraft were lost due to anti-aircraft fire and operational issues. Despite airborne opposition and unfavorable flying conditions, the mission achieved its objectives, significantly diminishing the Bonins' effectiveness as an aircraft staging area. After the afternoon raid, the two task groups headed south as per Mitscher's orders to position themselves off Saipan by the 18th. This operation proved successful, as around 100 Japanese aircraft that could have intervened in the impending battle were eliminated. On the morning of June 15, while the landing craft were lowered and began the complex assembly process, the Americans carried the last air and naval preparations on Saipan. At 8:12am, Turner finally gave the landing signal and the amtracs began to approach their assigned beaches under the cover of the LCI gunboats' rockets. Enemy fire was light until the LVTs reached the coral reef's lip, at which time artillery, mortar, and machine-gun fire increased drastically. Some control craft had drifted slightly to the north and coupled with a stronger than expected current some 2nd Marine Division units landed north of their intended beaches. The UDTs had not experienced this as the sea conditions and tides were different when they went to work the day before. Once over the reef lip some 400yds from shore, the swell made stationkeeping even more problematic. The swell increased through the day, making unloading and transfer difficult as well as capsizing landing craft and LVTs. The 2nd Marine Division had 31 of its 68 amphibian tanks hit or mechanically disabled before reaching the Tractor Control Line ashore. Nonetheless, in 20 minutes, some 8000 assault troops were ashore despite the strong Japanese artillery fire, albeit suffering heavy casualties. On the Red Beaches, Colonel James Riseley's 2nd and 3rd Battalions encountered stiff resistance from Hill 500 but still managed to secure a shallow beachhead, subsequently repelling a savage counterattack as the reserve 1st Battalion was landing. To the south, as Colonel Clarence Wallace's 2nd and 3rd Battalions had both come ashore on Green Beach 1, the 2nd would be sent south to seize Afetna Point while the 3rd advanced inland. Afetna Point was important for another reason than just denying the Japanese excellent positions for enfilading the American landing craft. Possession of the point would make Beach Green 3 available for landing of the tanks of both divisions. Because of an open channel off Green 3, LCM's carrying tanks could proceed directly to the beach without crossing the troublesome reef that fenced all other beaches. Once through the channel the craft could either move straight into Green 3 or fan out to the north or south and put the tanks ashore, wherever it was desired. The same channel could also be used for logistical purposes after tanks were ashore. As a result, the 2nd met strong resistance, progressing slowly and costly against a determined foe while the 1st Battalion was being landed; yet the 3rd managed to push inland against weaker resistance, finally halting about 900 yards from the O-1 Line. In the afternoon, the reserve 1st Battalion, 29th Marines was also landed and employed to fill the gap between Wallace's battalions. It's also important to note that the 4th Company, 9th Tank Regiment conducted piecemeal counterattacks on either side of Charan Kanoa with 14 tanks, losing all but three. Meanwhile, along the Blue Beaches, the 23rd Marines led by Colonel Louis Jones made headway through Charan Kanoa, facing occasional gunfire, and successfully advanced to the O-1 Line. Despite encountering challenging terrain and harassment from Japanese riflemen, their amtracs managed to secure Mount Fina Susu, albeit with significant losses. Upon reaching the O-1 Line, heavy artillery fire pinned down the Americans. Because of the concentrated mortar and artillery fire which had struck his troops throughout the day, the commander of the 23rd Marines, Colonel Jones estimated that the light forces on the O-1 line would be incapable of holding there for the night without being virtually annihilated. In addition, the regimental commander felt that “pulling back would allow our artillery and air to bring fire on the Jap batteries a short distance inland, better contact could be obtained on the right, and the exposed left flank could be better protected.” Dillon's 2nd Battalion and Cosgrove's 3rd Battalion were therefore ordered to establish a defense line generally 800 yards west of O-1 for the night. This involved a withdrawal under cover of darkness of the meager forces on O-1.Concerned about the exposed left flank between the 2nd and 4th Divisions, Colonel Jones directed the reserve 1st Battalion to fill the dangerous gap. Further south, Colonel Merton Batchelder's 1st Battalion faced fierce resistance from Agingan Point, repelling enemy counterattacks with the aid of air and naval support. Reinforcement from 3rd Battalion units allowed partial clearance of Japanese positions. The 2nd Battalion to the left encountered enemy concentrations but benefited from the heroic actions of the LVTAs, enabling them to advance inland.  About half the assault Marines of this battalion were carried to a railroad embankment prior to debarkation from LVTs. Until it reached there, the battalion was subjected to frontal small-arms fire from Japanese behind the embankment. There the small arms fire ceased, but mortar and artillery fire continued. From the direction of Nafutan Point came shells which burst overhead, apparently from antiaircraft guns depressed to fire at ground targets. And, as if troubles to the front and flanks were not enough, two enemy mortars began firing on the Marines from positions 500 yards to the rear. Before a request for assistance could be made, friendly planes spotted the mortars, attacked, and silenced them. Much credit for the 2nd Battalion's success in pushing inland goes to the LVT(A)s of the Army's 708th Amphibian Tank Battalion, which, according to the Marine battalion commander, Lieutenant Colonel Hudson, took “more than their share of punishment diverted enemy attention from the amphibious tractors carrying troops ... I shall always re- member the excellent support given to my battalion by the Army LVT(A)'s”. Meanwhile, Marine tanks and artillery, under Turner's command, were successfully brought ashore, although some tanks and howitzers from the 4th Division were lost to fire and reef obstacles. By late afternoon, the bulk of the reserve 2nd and 24th Marines were landed, with the latter sustaining 400 casualties from accidents and artillery fire.  Lieutenant Colonel Alexander A. Vandergrift,  therefore, deployed his battalion and moved it on foot to an assembly area some 700 yards inland from Yellow 1, near a fork in the rail- road tracks. The men had no more than taken entrenching tools in hand when a barrage of well-directed Japanese artillery fire engulfed them. After the day's casualties were totalled, it was discovered that the unit had suffered heavily, though it had yet to enter the front line fighting: 25 killed, 72 wounded, 39 missing, mostly those lost on board the overturned LVTs. Other battalions had suffered more heavily, but the real significance of these figures lies in the fact that the 3rd Battalion did not arrive on the beach until 5:27pm. However, by nightfall, only the 25th Marines had reached the O-1 Line, constituting less than one-third of the planned beachhead, with casualties exceeding 2000 during the landings. After half an hour of the Saipan Invasion, Admiral Toyoda announced Operation A-Go proclaiming "The fate of the Empire rests upon this single battle. Every man is expected to do his utmost." The same Nelsonia words used by Admiral Togo at the legendary battle of Tsushima in 1905. He had created that statement taking Nelson's signal "England expects that every man will do his duty" at the Battle of Trafalgar. Toyoda planned to destroy the American fleet. Admiral Ugaki's battleships were to rendezvous with Admiral Ozawa's carriers in the Philippine Sea to attack the American fleet around Saipan. At 6:35pm on June 15th, the submarine Flying Fish reported that a force of Japanese battleships and at least three carriers had passed through the San Bernardino Strait. The submarines were under orders to report sightings before attempting to attack, so Flying Fish waited until nightfall, then surfaced to radio in its report. The sighting report from the submarine was monitored by the Japanese who knew they had been located. About one hour later, submarine Seahorse also sighted a Japanese force 200nm east-southeast of Surigao Strait. The exact composition of this force was unclear, but was thought to be Yamato and Musashi plus escorts. The submarine was not able to send a signal of its sighting until 4:00am on June 16th. Additionally, throughout the 15th, a growing body of signals intelligence indicated that a major IJN fleet action was underway.   Receiving this intelligence on the enemy's intentions, Admiral Spruance made the decision to halt the Guam landings, instructed for unloading at Saipan to continue until June 17th only, and tasked Turner with preparing for an imminent decisive battle. However, that's a podcast for next week. General Saito was on the verge of initiating his counterattack. Throughout the night, there were various probes and minor counterattacks that were swiftly repelled. Nonetheless, the 2nd Battalion, 6th Marines faced a major assault at 3am when a Japanese bugler 75 sounded a loud, clear call on the tense night air, and with a waving of flags, loud screams, and a brandishing of swords, the attack was launched. Added to the din were the shouts of the 6th Marines for “flares.” Illumination shells, fired by the three destroyers, Halsey Powell, Coghlan and Monssen of Task Unit 52.17.3 began to burst over the area, clearly revealing the Japanese attackers. Men of Companies F and I opened with accurate, devastating lire. California joined with a tornado of salvos in front of the Marine lines. The Japanese, stripped by the illumination of the advantage afforded by their superior knowledge of the terrain (so helpful in night lighting), dropped before the Marines' fusillade. As Japanese fell, others replaced them in the determined onslaught.  At one point in the fighting, it was feared that the enemy had effected a penetration of Company I, along the coastal road. To contain this Japanese prong, Colonel Riseley shifted one company of the 3rd Battalion, 2nd Marines, company L, into the area. The light of morning, however, revealed that no penetration had been made and that Company B had, in effect, formed a secondary line. By 5:45, the Japanese pressure reached a peak; two Marine 37mm guns near the beach were knocked out and their crews forced back. Although the main positions held, a false report reached the 6th Marines' command post to the effect that Company F's lines along the beach had been forced back about 50 yards. This erroneous report, probably fostered by the withdrawal of the two 37mm crews and the infiltration of small enemy groups to the regimental command post, had no basis in fact. Fortunately, five medium tanks arrived in the nick of time to halt the Japanese advance and thwart their assault. With the dawn of June 16th, the battlefield revealed approximately 700 enemy casualties strewn across the terrain. Meanwhile, as the 6th Marines consolidated their position and the remaining 2nd Marines landed, Colonel Wallace's 2nd Battalion resumed its offensive against Afetna Point, successfully eliminating enemy resistance and bridging the gap between Marine divisions. Further south, General Schmidt led his three regiments towards Fina Susu Ridge in a coordinated advance. Despite the 23rd Marines encountering formidable enemy opposition in the mountains and marshes, the 25th Marines managed to clear Agingan Point and extend beyond the beachhead, while Colonel Franklin Hart's 24th Marines in the center made significant progress towards the O-1 line despite facing fierce resistance. Meanwhile,  Spruance opted to withdraw the unnecessary transports in anticipation of the impending naval confrontation. General Smith made the decision to deploy the reserve 27th Division for landing. Consequently, Major-General Ralph Smith's units, excluding the 106th Regiment, commenced landing operations in the afternoon. However, during the night, Saito initiated his second significant counterattack. The Japanese 136th Infantry (two battalions) and 9th Tank Regiments were to counterattack the center of the 2nd Marine Division sector at 17.00hrs. Yokosuka 1st SNLF was to attack the north flank from the Garapan area and “cooperate with the tanks to annihilate the enemy's front line and advance towards Oreai (Charan Kanoa) Airfield.” Vice Admiral Nagumo, Commander Central Pacific Fleet and 5th Base Force, was ordered to command the attack personally. The units were still disorganized from the previous night's action and the attack was not launched until 03.30hrs. The objective of the Japanese attack was modest enough: the Saipan radio station, a short 400 yards behind the 6th Marines' lines. This objective, in itself, represented a revision in Saito's pre-D-Day policy to “destroy the enemy, during the night, at the water's edge." Seizure of the radio station area would still leave 500—600 yards distance to the water. This resulted in an estimated 44 tanks and 500 infantry attacking the positions of the 1st Battalion, 6th Marines, and 2nd Battalion, 2nd Marines. The attack began at about 3:30, and the brunt struck Lieutenant Colonel Jones' 1st Battalion, 6th Marines,  and to a lesser extent the 2nd Battalion, 2nd Marines. The tanks advanced in groups of four or five with Japanese soldiers clinging to them. Poor and ineffective tactics reflected the inadequacy of Saito's order; some tanks cruised about in an aimless fashion, some bogged down in the swampy ground, some made an effort to break through the lines, still others stopped to let off their pugnacious passengers. The Japanese suffered heavy losses, including approximately 24 tanks and 300 men, while inflicting only 97 casualties. By the end of the initial two days, American forces had incurred roughly 3,500 casualties, approximately 20% of the total for the operation. Concurrently, General Obata had instructed the commanders of Guam, Truk, and Tinian to send reinforcements to Saipan via barge, although only two companies managed to land on Rota by June 22nd. On June 17th, the American offensive resumed, yielding significant progress, partly due to artillery support. General Watson expanded the beachhead area nearly twofold, with the 3rd Battalion, 2nd Marines, and 2nd Battalion, 6th Marines advancing northward to within 1000 yards of Garapan. Meanwhile, Colonel Riseley's remaining units secured the base of Mount Tipo Pale along the O-2 Line, and Colonel Wallace's Marines maneuvered around Lake Susupe to reach the O-1 Line.  To the south, the 25th Marines led by Colonel Batchelder advanced in a column formation, making headway towards Aslito Airfield by approximately 1500 yards. They managed to secure advantageous high ground along the O-2 Line on the northwest perimeter of the airfield. Simultaneously, Colonel Gerard Kelley's 165th Regiment also moved towards Aslito, making gains of around 1200 yards despite encountering light resistance. By 2:00pm, the 2nd Battalion had moved to the southwest edge of Aslito Airfield, while the 1st Battalion had become involved in a fight for a dominating ridge line extending from the airfield's southwest corner to the vicinity of Cape Obiam. Infantrymen of the 1st Battalion led by Major James H. Mahoney inched slowly up the steep and open slope, so that, by 5:00pm, a toehold had been gained at the crest. At this point, however, the Japanese launched a counterattack and drove the soldiers back off the hill. Reorganizing and reforming the lines at the base of the ridge, the battalion dug in for the night. Amid heavy artillery bombardment, the 24th Marines reached and secured the O-2 Line, consolidating their position on the ridge. In contrast, progress for the 23rd Marines to the south of Lake Susupe's swamps was minimal as they faced entrenched enemy forces. With the expansion of the beachhead, additional Corps units landed, including General Smith's command post, advance parties of the Corps artillery, and Colonel Leonard Bishop's 105th Regiment. During the night, Saito launched another counterattack, deploying the 1st Battalion of the 18th Regiment to stage a counterlanding on the Marine beachhead. However, the intervention of LCI gunboats and Marine artillery thwarted the attempt, driving the enemy back to Tanapag Harbor. Shifting focus to the Battle of Biak, General Fuller's forces had previously seized Mokmer but now found themselves encircled by Colonel Kuzume's formidable positions in the East and West Caves. Despite progress in repairing the airdrome, intense Japanese fire prevented its use until the enemy cave strongholds were neutralized. In contrast, the Japanese had attempted twice to reinforce Biak but failed both times. However, Operation KON drew the Allied forces away from the island, allowing barges to discreetly bring reinforcements from the southwest to Biak without naval protection. As a result, most of the 2nd Battalion, 221st Regiment was successfully transported to the island by June 12, and General Numata was evacuated during this period. Following this success, the 2nd Battalion, 219th Regiment was loaded onto barges in Sorong and transported to Biak via Noemfoor Island, arriving successfully by June 16. Meanwhile, as Colonel Newman's 186th Regiment secured its positions on June 8, Fuller decided to send the rest of Colonel Haney's 162nd Regiment to the recently secured Parai Jetty. With the Ibdi Pocket contained, Haney made significant progress toward Mokmer. The next day, Haney continued his advance, successfully connecting with his 2nd Battalion despite heavy mortar fire. Newman faced multiple night counterattacks and was unable to expand his beachhead. On June 10th the 1st Battalion, 186th Infantry, sent two companies to the point on the low ridge where Company B had been halted the previous afternoon. Despite artillery support, the two units could make little progress and were themselves pinned down about 10:30. Japanese rifle and mortar fire was silenced by the 1st Battalion's 60-mm. mortars, but the Japanese continued to pour machine gun fire from a number of bunkers and pillboxes which proved impervious to bazooka and 75-mm. tank fire. The units withdrew while more artillery fire was placed along the low ridge. On the east flank, enemy fire from the East Caves had died down, and the 2nd Battalion, 162nd Infantry, was able to move on eastward. On that same day, as Admiral Toyoda ordered a third attempt for Operation KON, Admiral Ugaki's battleships were directed to support Admiral Sakonju's convoy, assembling at Batjan Island on June 11th. The start of the Marianas bombardment on that day forced Toyoda to suspend the operation and prepare for A-Go, ordering the battleships to rejoin Admiral Ozawa's command and sending the 23rd Air Flotilla to Palau to reinforce Admiral Kakuta's 1st Air Fleet. With two regiments now stationed at Mokmer, Fuller decided to resume his offensive by advancing westward towards Borokoe and northward towards the low ridge. The 163rd Regiment was assigned to hold the Ibdi Pocket and the East Caves. On June 11, the assault battalions began moving to the starting point in front of Menoebaboe. Haney's forces faced strong resistance and only reached the starting point by afternoon. The Americans then crossed the starting line, encountering minimal resistance along the coast but significant opposition on the low ridge. As a result, three battalions reached the 1st Phase Line, while Haney's 3rd Battalion couldn't move beyond the starting line. This forced Newman to pause his advance temporarily until Haney cleared enemy positions on the ridge. On June 12, Haney's 2nd and 3rd Battalions advanced on the low ridge. The 3rd Battalion fought doggedly forward during the afternoon, discovering an ever increasing number of Japanese pillboxes, bunkers, and hasty automatic weapons and rifle emplacements of all kinds, manned by 1st Battalion and 5th Company, 222nd Regiment. Dusk found the unit still some 100 yards short of the line of departure and about 1,300 yards east of the 2nd Battalion. The 1st Battalion, taking no part in the action during the day, moved forward to Mokmer Drome from Parai. Due to heavy Japanese resistance, their progress was slow. The next day, the Americans continued their attacks against enemy positions, but they couldn't close the gap between Haney's 2nd and 3rd Battalions. Fuller concluded that his troops were fatigued and that the Japanese had likely received reinforcements, so he requested an additional infantry regiment. General Krueger planned to send the 34th Regiment from Hollandia, scheduled to arrive on June 18. In response to recent setbacks, General Doe ordered the 1st Battalions of the 162nd and 186th Regiments to support the 3rd Battalion on June 14. These battalions positioned themselves on the terrace above the low ridge, facing the main Japanese strongpoint directly. In response to the enemy's aggressive action, Colonel Kuzume initiated a tank-infantry counterattack with his remaining forces to push back the enemy battalions, inflicting heavy casualties and disrupting the American advance. Despite this, the Americans managed to regroup and establish a continuous line, while the Japanese sustained significant losses and were unable to execute any further major counterattacks for the rest of the battle. Generals Krueger and MacArthur were unhappy with the progress of operations in Biak, which led them to relieve Fuller of his duties as task force commander. Whether General Fuller's relief as commander of the HURRICANE Task Force was entirely justifiable is a question which cannot be answered categorically. At the time of his relief, the task force had seized Mokmer Drome. Patrols sent westward to Borokoe and Sorido Dromes had found no enemy at those two fields, and General Fuller knew they could be occupied with ease. But he had not sent more troops beyond Mokmer Drome because he believed it more important to secure an overland line of communications to that field and to clear the low ridge so that repair work could continue and at least one strip could be put in service. By June 14th it was only a question of time before the West Caves area and the low ridge would be secured. Indeed, General Eichelberger, who took 3.5 days to acquaint himself with the situation at Biak, drew up new attack plans according to which the 162nd and 186th Infantry Regiments were to be employed in the same area and in much the same manner as General Fuller had been using them. General Eichelberger realized, as had General Fuller, that Borokoe and Sorido Dromes would be no safer than Mokmer Drome as long as the Japanese held the low ridge and West Caves positions. But, in the last analysis, the mission of the HURRICANE Task Force, quick seizure and rehabilitation of the Biak fields, had not been accomplished by mid-June. General Eichelberger then took command of Operation Hurricane on June 15. Although the initial plan was to keep Fuller in charge of the 41st Division, he felt unjustly treated and requested to be relieved of this role as well. Consequently, Doe was promoted to command the 41st Division in Fuller's place. I would like to take this time to remind you all that this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Please go subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry after that, give my personal channel a look over at The Pacific War Channel at Youtube, it would mean a lot to me. The invasion of Saipan was seeing tremendous casualties for both the Japanese and Americans. By the end of the initial two days, American forces had incurred roughly 3,500 casualties, approximately 20% of the total for the operation. This island was not going to be a cake walk by any measure, it would be paid for in blood.  

Last time we spoke about landings at Biak. General Fuller unleashed a amphibious assault against Biak that faced countless hurdles. The Hurricane Task force encountered a lot of terrain issues at Humboldt bay, leading to logistical headaches. Despite the disorganization, they shipped off and landed, forming a beachhead. Colonel Kuzume and his men were caught with their pants down, units were scattered all over the place. The first tank battle of the Southwest Pacific occurred, seeing American Shermans absolutely desolate Japanese Type 95's. General Fuller planned to consolidate his troops at Ibdi and Bosnek while reinforcements arrived, but the Japanese continuously lobbed surprise night attacks to horrible effect. Over in the Burma front, Mutaguchi's operation continued to unravel as his subordinate officers disregarded his orders and performed their own withdrawals. As Mutaguchi relieved men of command and replaced them, General Slim finally reopened the Imphal-Kohima road spelling doom for the Japanese. This episode is the Fall of Mogaung Welcome to the Pacific War Podcast Week by Week, I am your dutiful host Craig Watson. But, before we start I want to also remind you this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Perhaps you want to learn more about world war two? Kings and Generals have an assortment of episodes on world war two and much more  so go give them a look over on Youtube. So please subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry for some more history related content, over on my channel, the Pacific War Channel you can find a few videos all the way from the Opium Wars of the 1800's until the end of the Pacific War in 1945.  As General Vinegar Joe unleashed what he believed to be a masterstroke against Myitkyina, it soon turned out to be an absolute gruesome struggle. As we last saw, General Stilwell's men had begun a long and difficult siege of Myitkyina. The 22nd and 38th Chinese Divisions were now pinned down by General Tanaka's battered 18th Division south of Kamaing. To the south Brigadier Calvert's Chindits began a battle for Mogaung, which forced General Takeda's 53rd Division to suspend the relief of Myitkyina and rush back to reinforce the town. Though the Mogaung Garrison and the 1st Battalion, 128th Regiment to the north had been effectively destroyed by the Chindit and Chinese attacks, the Japanese had been able to maintain their hold on Mogaung by mid-June. To the east, General Wei's Y Force had opened a new offensive on Yunnan, gradually pushing Colonel Kurashige's 148th Regiment to Tengchong but failing to seize Longling against the tenacious resistance of General Matsuyama's forces. Along the Kamaing-Mogaung front, by late June, Tanaka had been able to assemble most of his depleted command at Lakatkawng, determined to keep the blockade on the Hukawng Valley. His main aim was to clear the Seton roadblock, which had been recently reinforced with General Sun's 113th Regiment; but once again, his attacks would fail to dislodge the tenacious Chinese defenders. Yet upon receiving orders from the 33rd Army commander to withdraw, General Tanaka reported that the 18th Division could continue to hold in the Kamaing area. This statement, inspired by Colonel Ohgoshi, the chief of staff, proved to be unwarranted optimism. The 18th had staged a desperate resistance in the vicinity of Kamaing for about a month and, for most of the period, had only 80 men for each mile of front. Supplies of ammunition and food were very low with only about 1400 rounds of rifle ammunition per day for the entire Division and 60 rounds per machine gun. The daily ration was about 2.5 ounces of rice per man. On receipt of the Army's message to withdraw, Colonel Ohgoshi had advised the Division commander that further resistance in the area was possible, but had not made it clear that this was his personal belief and did not reflect the opinions of the rest of the Division staff. Within a day or two the commander became aware of the fact that the other staff officers were convinced that further resistance in the Kamaing area was impossible. He therefore forwarded to the 33rd Army a revised report of the Division's actual situation. Upon receipt of the second message, on June 27 the Army directed the Division to retire to the Sahmaw sector. Tanaka believed he needed to stand his ground while the 53rd Division pushed aside the Seton Block and reopened his line of communications. Thus, he elected to continue to resist the attacks from the north while he himself attacked Seton for three more days; but failing to make any progress, he would finally comply with his orders to withdraw to the hills north of Sahmaw in early July.  While the 4th and 146th Regiments performed covering attacks, the remnants of the 55th and 56th Regiments destroyed their artillery and heavy equipment, and withdrew along an escape trail cut through the forest west of the Seton roadblock. On 2 July the 18th Division began its withdrawal, utilizing an obscure trail that ran directly south from Kamaing. Crossing the mountain range west of Seton, the Division completed its concentration near Sahmaw about 7 July. The Hukawng Operation was then considered concluded, ending a campaign that had been a miserable failure and had cost almost 8,000 casualties. By July 15th, the 18th Division would finally assemble in the Indaw area. Though only 3000 men from the elite 18th Division would survive the Hukawng Valley Campaign, Tanaka had effectively managed to keep intact the blockade to China for another year, something that would have profound repercussions later on in Chinese history. Further south, the 114th Regiment finally arrived at Gurkhaywa on June 16th, ready to reinforce the Chindits; yet Takeda had also brought most of his troops back to Mogaung, subsequently starting a deadly shelling of the Chindit positions. By when June 15th arrived, the Chinese still had not appeared, and Calvert pulled his troops back towards the bridge. At that moment, remarkable news arrived: The Japanese were abandoning their positions along the river. Calvert was exuberant. This meant he could move out of his bridgehead perhaps capture the town. Certainly, it meant a reduction of the shelling which was claiming at least 15 of his men a day. Yet, when the shelling did not die down and it quickly became apparent that Takeda was merely redeploying his troops along the railway, to get them out of flooding in low-lying areas. Chindit recce teams reported the area from the train station, in the heart of the town to the Mogaung Railway Bridge, further north, was heavily defended with eight bunkers dominating the landscape. Shelling from the village of Naungkyaiktaw, astride the road to Mogaung, set between fields of paddy, was persistent. Naungkyaiktaw had to be captured. Calvert estimated the village was held by a hundred Japanese. Because of this, on June 18th, Calvert ordered his forces to attack the apparent Japanese artillery encampment at Naungkyaiktaw after a heavy air and mortar bombardment. His troops outnumbered them, but unwilling to suffer needless casualties, Calvert directed the American fighter-bombers against the village, which was bombarded on the night of the 17th. Half an hour before dawn on the 18th, the Chindit mortars pummeled the place with 400 rounds for good measure. Calvert then sent in his assault force. Among the attackers was a company of 70 men from King's Liverpool led by Major Fred Reeman. This was a company that had stayed on with the 77th Brigade after the rest of the battalion had been transferred to the 111th Brigade. They were joined by 12 men of Blaine's Detachment, once evacuated to India but since returned, this time armed with about a dozen flamethrowers.  In the darkness, Blaine's Detachment was told to advance in front of the company of King's, and to “turn the fucking lights on.” As the detachment began to hurl flames far and wide, the Chindits behind them began cheering. The men had been told that the village had many bunkers, but never saw any at first. The scene soon turned fantastic. They went through the entire village “with twenty or thirty yards of flame shooting out in front.” They soon found the bunkers. The Japanese became crazed with fear especially after the British began yelling “put out the fucking lights,” and turning the flamethrowers their way. Many Japanese fled the bunkers, joined by those outside. They fled through the paddy fields, making for the railway station 400 yards away. Calvert's machine gunners had been waiting and blazed away, killing at least forty. Meanwhile, the rest of Fusiliers and the Kings walked up the paddy, picking off Japanese hiding or trying to crawl away in the ditches. Calvert, his mobile brigade-major Brash and his orderly Lance-Corporal Young decided to join the mop-up, shooting at Japanese while standing on chairs, as more Fusiliers began clearing the last of the bunkers, hurling grenades into them and blasting the insides with flamethrowers. As twilight set in that day, the most peculiar thing happened. The Fusiliers were cooking an evening meal in their newly-won positions, when a weary, seven-man patrol walked into their billet and began to take off their kit. The Fusiliers who looked up casually, noticed to their horror, that the new men were Japanese. The Japanese, for their part, had not noticed. The Fusiliers lunged for their weapons and opened fire. The Japanese patrol did not survive. In all, Calvert estimated that his troops had killed about 70 Japanese in the capture and holding of Naungkyaiktaw, while suffering 16 killed and 38 wounded. Major Reeman's King's company had become reduced to a platoon.  Calvert was considerably cheered on the evening of the 18th, when the much promised Chinese reinforcements finally arrived, guided over the river in motorized ranger boats by a towering Chindit officer, Captain Andrew. This was the 1st Battalion of the Chinese 114th Regiment led by Major P'ang, which quickly deployed in the positions pointed out by Calvert but left the Chindits a little flummoxed when they proclaimed that they were in no particular hurry to fight as they had been fighting for years. On the following day, another battalion of Chinese arrived under the personal command of the regimental leader, Colonel Li Hung, as did a battery of 75mm pack howitzers  the “6th Battery” under US Major Wayne Cook. The Chinese quickly assumed the defense of Mahaung, prompting an American liaison officer with the Chinese to send a press release that the Chinese had “captured” the village, which embarrassed Li. Cook's battery was deployed into position at Pinhmi village began operations on the 20th, hurling fire against the Japanese positions as the Chinese infantrymen consolidated their positions. Meantime, elements of the Chinese 113th Infantry, operating five miles north of Mogaung, surrounded a Japanese company, while Cook's guns hammered them. Fifty Japanese died from first blast alone. The Chinese finished off those who survived.The assault, was so ferocious that all the bunkers were overrun The reinforcements heartened Calvert for his own brigade was now a shell of its former self. The Lancashire Fusiliers and the King's Liverpool had only 110 men, the South Staffords had 180 and the Gurkha Rifles had 230. He planned a fresh advance, this time aiming for the hamlet of Natgyigon, on Mogaung's right flank, near the river. This area, Calvert believed, was the “key to Mogaung.” For the time, he chose the early hours of darkness on June 23rd a day which would go down in the annals of the 77th Brigade as the “stuff of legends.” The plan called for a mortar barrage of 1,000 bombs, in addition to shelling from the 75mm guns to cover the advance of the Chindits across the open ground towards Natgyigon. The Gurkhas were to move on the right, with the South Staffords on the left. Blaine's Detachment and the Lancashire Fusiliers were in reserve. The objective was to capture the entire stretch of ground from the Mogaung Bridge to the train station, the latter of which the Chinese were to secure. Once the area was in Allied hands, the troops were to dig in while the reserve troops mopped up. In addition, Allied aircraft were to bombard the area before the start of the assault, which itself was timed to launch at about 3.10 am. In the dark, section commanders could be heard telling their men: “We attack Mogaung tonight and once we've taken it the Brigadier says we are through!” Later, during the attack, Calvert discovered the Chinese infantry had not captured the all-important train station, even as their American liaison officer insisted that they had. Calvert angrily pointed out that no, the Chinese had not, because enemy fire from that direction continued to pick off his men at the railway embankment. The Gurkhas, moving along a wide right flank along the banks of the Mogaung River, headed for the railway bridge. Approaching the bridge, they came under heavy fire. Captain Allmand, by now suffering from trench foot as were most of the troops, moved forward to silence a machine gun firing on his men. He could barely run because of his affliction but advancing through the mud, he hurled grenades at the Japanese position. A burst of gunfire plunged into him. He fell, badly wounded. One of his Gurkhas, Sergeant. Tilbir Gurung pulled Allmand and another wounded NCO to safety. For this, Gurung was to get a Military Medal. Allmand's own valor was to be recognized by a Victoria Cross. The South Staffords swept into Mogaung town. Resistance was heavy. Lt Durant of the South Stafford deployed his machineguns to rake Japanese positions with fire. Meantime, the flamethrowers were brought up. As they moved up past Durant's positions, a shell burst exploded one, setting the man wielding it on fire. The man screamed and somehow shook off the flamethrowing unit from his back. Durant and some of his men rushed forward and rolled him into water in a nearby ditch. The Japanese had dug-in beneath the ruins of a brick house from where they were stubbornly holding the Staffords at bay. The rest of the flamethrowers moved in and sprayed the building. One Japanese, his clothes ablaze, leapt from his positions and tried to make a run for it. A scythe of gunfire cut him down. The rest valiantly held their positions and were burned to cinder. The Staffords, mopping up the, found the Japanese officer. He had shot himself with his revolver. The Japanese had entrenched themselves at a strategically important building known as the Red House, which was well-protected with machine-gun nests. The advancing Gurkhas consequently ran smack into this killing zone, getting caught in a murderous crossfire and suffering heavy casualties. In response, Calvert threw his reserves into the fray and the Chindits also began to pummel the Japanese positions with mortars and machine-guns, which allowed the infantry to reach the all-important train station. Inflicting some 120 casualties and losing 60 dead and over 100 wounded, the Chindits then successfully captured all their objectives by noon. For the rest of the day, heavy fighting would continue as the Chindits dug in on their gained positions; but during the night, the Japanese would finally pull out, leaving the town to the shattered remnants of the 77th Brigade. Mopping up then continued until June 27, when Mogaung was declared void of Japanese. Though this was the first major town to be recaptured in Burma, Calvert lost over 250 killed and 500 wounded at Mogaung, which was more than any Chindit formation was prepared to take. This was also a bittersweet victory for Calvert because Stilwell would claim that the town had been taken by his Chinese troops, even though the Chindits had done most of the fighting. Stilwill wrote in his diary on June 27th “Good news from Mogaung, We have it!” Then came a remarkable broadcast from Stilwell's headquarters via the BBC “The Chinese had captured Mogaung”. There was no mention of the Chindits. Calvert was incensed. Colonel Li was appalled and apologized profusely. “If anyone has taken Mogaung it is your Brigade and we all admire the bravery of your soldiers.” Calvert, his anger against Stilwell unmitigated, sent a message to US headquarters  “Chinese reported taking Mogaung. My Brigade now taking umbrage” this prompted Stilwell's staff to scour the maps for the location of Umbrage. Meantime, congratulations poured in from Lentaigne, from “Scottie” Scott, from John Masters, and the other brigade commanders. Among the lot, there was one, from Derek Tulloch, which struck Calvert's heart the most: “Wingate would have been proud of you.” After this defeat, and learning of the concurrent withdrawal of the 18th Division, Takeda's 53rd Division would withdraw to the Sahmaw River line in early July, where it was also reinforced with the recently-arrived 119th Regiment.  Meanwhile over at Myitkyina, General Boatner had to order a stop to the attacks after June 18th because of the heavy casualties. For the time being, tunneling would be used to close with the enemy. On June 25th, however, Boatner would have a severe recurrence of malaria that would force him to abandon the frontlines. This led General Stilwell to appoint Brigadier-General Theodore Wessels in command of the Myitkyina Task Force on June 26th. Luckily for Wessels the situation started to improve after the fall of Mogaung, as Chinese troops there could now move up the railroad to connect with Wessels' forces. This removed the recurrent menace of a Japanese drive from Mogaung, guaranteed reinforcements and the opening of a ground line of communications, and further eliminated one of General Minakami's two bases from which supplies had trickled into the Japanese perimeter. Despite this, the only gains in the week of June 25th were a few hundred yards taken by the 150th Regiment and the 236th Engineers. Alongside this, Stilwell ordered the 1st Battalion, 42nd Regiment to penetrate through the Japanese positions towards Sitapur on June 28. They would drive deep into the Japanese defense system, leading Stilwell to hope this was the turning point; on receiving Japanese fire, it halted and dug in. Air supply was necessary.In response, Wessels dispatched some Marauder reinforcements. F Company, unaware it had lost its way and under an inexperienced commander, proceeded with a small point almost directly ahead of the marching column. The company commander at the head of the point met a small group of Orientals whom he took to be Chinese and who greeted him affably. The strangers then suggested he and his party lay aside their guns. At this point the commander realized that he had been ambushed and gave the alarm. The Japanese machine guns opened on his trapped column, inflicting heavy casualties. Some of his men made their way back to the Allied lines, but the company was never reconstituted and was broken up and distributed among the rest of Galahad. For his constant gallantry during a stubborn eight-hour rear-guard action, which permitted the survivors to extricate themselves from ambush, Private first class. Anthony Firenze of New Galahad received the Distinguished Service Cross. Wessels then planning to launch a set-piece attack to capture a stretch of the Sumprabum Road.  Over in the Yunnan front, Colonel Matsui's 113th Regiment had successfully relieved the pressure from Longling by mid-June. General Matsuyama further ordered him to maintain the offensive while he continued to reorganize his forces. Though Matsui managed to seize the Tiechanghe pass on June 21st, most of his attacks would end up in nothing. In the north, the 20th Army Group launched simultaneous attacks against Qianshuang and Gudong on June 18th. This finally forcing the Japanese to retreat in disorder towards Tengchong by June 22nd. With the fall of Qianshuang, the Japanese had been forced to abandon the upper Shweli valley, and were now moving in some disorder toward Tengchong over three excellent trails. In Qianshuang, they left behind large quantities of ammunition and a few pieces of artillery, suggesting a disorganized withdrawal. 150 dead Japanese were found in Qianshuang itself; more than 300 Chinese gave their lives for the village. South of Qianshuang, the Japanese hastily destroyed their pontoon bridge to slow the Chinese pursuit. On reaching the Qianshuang-Baifen-Gudong line, the 20th Army Group had wrested 4000 square miles from Japanese control in forty days of fighting. The advance had been made over the precipitous ranges of the Kaolikung Mountains in an almost constant rain, a downpour sometimes heavy, sometimes light, rarely abating, and always turning to fog and sleet in the higher altitudes. More than 150 coolie supply porters fell to their deaths from the narrow, slippery trails that snaked precariously over the mountains. On June 25th General Wei received a personal order from Generalissimo Chiang Kai-Shek to take Teng-chung. A few days later, th 20th Group Army, though delayed by the need to rebuild bridges over each of the swift mountain streams that crossed its advance, had pressed the Japanese rear guards back to the hills that surrounded Teng-chung at a distance of two to three miles from the formidable walled town itself. Meanwhile, the Fourteenth Air Force was trying to soften Teng-chung by daily attacks with bombs and machine-gun fire. The outer defenses of Teng-chung were pillboxes covering every avenue of approach, supported and covered by the 6,500-foot-high, fortified mountain peak of Lai-feng Shan, "The Place Where the Birds Come." Here were 600 or more Japanese with most of the garrison's artillery. Teng-chung itself was girdled by a massive wall of earth that in some places was forty feet high and sixty feet thick at the base, faced throughout with great slabs of stone. Chinese necromancers had carefully laid out the wall in a great square to cut the cardinal points of the compass. Each side had a gate, and each gate now had a Japanese command post, while Japanese machine guns and rifles swept the approaches to the wall, its face, and its parapets. Within the city were about 2,000 Japanese. In all, Colonel Kurashige, who had defended the Kaoli-kung mountains, had about 1,850 Japanese, a heavily reinforced battalion combat team built around the 2nd, 148th regiment. Kurashige's orders were to hold Teng-chung until the Chinese threat to Lung-ling passed Over at Longling, Matsui saw the arrival of some reinforcements on June 22nd that would allow him to continue his counteroffensive. Making repeated night and day attacks, the Japanese would be able to penetrate the enemy positions on June 24. Matsuyama then directed him to exploit towards Bengmiao and Huangcaobacum; yet a heavy raid by 24 B-25s and the arrival of the 1st Division would manage to halt the Japanese attempt to exploit their success, with Mitsui only securing the area northwest of Bengmiao by July 1st. The next day, Matsuyama then suspended the counteroffensive because of heavy casualties and he could see the enemy were strengthening their positions. In the meantime, Major Kanemitsu's Lameng Garrison was successfully holding off against a siege by three divisions since June 4th, though the Chinese would only launch unsuccessful attacks in regimental strength during this period; and to the southeast, the Pingda Garrison was also successfully repelling the small enemy attacks against them in spite of being cut off and disease-ridden. That is all for the Burma front today as we now need to head over to the Biak front. After the arrival of two battalions of the163rd Regiment for reinforcements, General Fuller planed a two-pronged attack against Mokmer Drome, with the 186th Regiment advancing west over the inland plateau while the 162nd Regiment resumed its attack west along the coast. On the morning of June 1, in preparation for the offensive, Colonel Newman's 3rd Battalion therefore left Bosnek and marched north over the coastal ridge, with the 2nd Battalion also moving from Opiaref to join them. By 11:00, both battalions successfully set up defensive perimeters; yet their preparations would be interrupted abruptly in the afternoon as Colonel Kuzume directed his 1st Battalion to attack the positions held by Company K. These Japanese, who were supported by machine guns and mortars emplaced northwest of the trail crossing, continued attacks until 5:00, when a platoon of Company K, by a flanking movement, forced their withdrawal northward. Company K and two platoons of the Antitank Company remained at the trail crossing for the night. Company I was moved forward to K's left and left rear, and Company L extended K's perimeter east along the main road toward the surveyed drome. Battalion headquarters and Company M stayed near the strip's western end. The 121st Field Artillery Battalion, the Cannon Company, the 2nd Battalion, regimental headquarters, the attached engineers, and the tanks remained near the center of the airfield.   Thankfully, the Americans would manage to repel the assaults and would ultimately force a Japanese withdrawal via a bold enveloping maneuver. But the Japanese would return after midnight. The first part of the night passed without incident, but at 3:30 the entire area held by the 3rd Battalion, 186th Infantry, flamed into action. About a company and a half of the 1st Battalion, 222nd Infantry, moved from the south against the semicircular perimeter held by Companies I, K, and L, having outflanked the 3rd Battalion on the west. Simultaneously, other elements of the 1st Battalion attacked from the northwest, attempting to drive a wedge between Companies L and K. Under the support of mortar and machine gun fire from both the northwest and southwest, the encircled Japanese desperately tried to fight their way north. Four hours of confused hand-to-hand fighting, marked by the use of bayonets, machetes, and grenades, ensued. At daylight a count revealed that 86 dead Japanese were within and around the 3rd Battalion's perimeter. The dead included the commander of the 1st Battalion, 222nd Infantry. Losses to the American unit were 3 men killed and 8 wounded.  After dealing with the threat, Newman resumed the westward advance at 9:00 on June 2nd. The 1st and 3rd Battalions, supported by five tanks and an antitank platoon, were to advance abreast, while the 2nd protected the right flank by patrolling north of the main road. The 121st Field Artillery Battalion was to provide continuous close support and was to displace forward with the infantry. Neither artillery nor air bombardment seems to have been provided for or delivered prior to the attack. However, both the 121st and 146th Field Artillery Battalions were registered on targets north and west of the 186th Infantry. Air support was available from Wakde Island upon call. The 1st Battalion, 186th Infantry broke camp at its beach defense area at 8:00 on June 2nd and moved north over the ridge to join the rest of the regiment. The 1st Battalion, 222nd Infantry, had made no serious attempt to stop the 186th Infantry's progress westward because the inland plateau was nearly indefensible and because the battalion would have been decimated in battle with the superior strength of the reinforced American regiment. The 1st Battalion was withdrawn from the surveyed drome area, initially in preparation for counterattack against the Bosnek beachhead. While no such counteroffensive was mounted, the withdrawal of the 1st Battalion at least had the advantage of keeping the unit intact. The American advance would thus be opposed by the 10th Company, 222nd Regiment; the 3rd Company, 36th Division Sea Transportation Unit and some other naval and engineer units.  The 1st and 3rd Battalions then advanced with two companies abreast against scattered but determined opposition from elements of the 1st Battalion, 222nd Infantry. Small enemy patrols aimed machine gun and rifle fire at the advancing American units and held their positions until killed or dispersed by tank or artillery fire. Most of the enemy parties were located on the north flank and apparently many of them had been driven westward out of the cave and garden area north of the surveyed drome by fire from the 121st Field Artillery Battalion, which destroyed Biak Detachment headquarters installations in that area. By nightfall the 186th Infantry had killed 96 Japanese and had itself lost 6 men killed and 10 wounded. The unit halted shortly after 1600 and began digging in at a point about 600 yards northeast of the day's objective. The advance had carried the regiment west until it was almost abreast and north of the 162nd Infantry, at the Ibdi Pocket. The latter had attempted to move west along the coast during the day, but it would be unable to dislodge the Japanese from the Ibdi Pocket, ultimately having to attach its 2nd Battalion to the 186th.  The addition of the 2nd Battalion, 162nd Regiment to the 186th Regiment helped to complicate the supply problems of the troops on the plateau. No water had yet been found inland. Heat and humidity were intense, and thick scrub growth, about 12 feet high, stopped any breezes. Despite the best efforts of Company B, 116th Engineers, the supply road could not be repaired fast enough to keep pace with the advancing infantrymen. Water had to be brought around from Bosnek via Opiaref to the forward units, and there were not enough water trailers nor 5-gallon cans available to supply all the water needed. At night each man received only one canteen of water for the next day, an inadequate amount under the conditions which prevailed inland. The water situation and the necessity for hauling all other supplies north through Opiaref did more to delay the 186th Regiment's progress westward than did the opposition of the 1st Battalion, 222nd Regiment. Meanwhile Kuzume's only support so far had been some air raids carried out by the depleted 23rd Air Flotilla and 7th Air Division. By late May, the 23rd Air Flotilla had only twelve fighters and six medium bombers at Sorong and the 7th Air Division had four large bombers, 20 medium bombers and three fighters. Both units threw what strength they could muster into attacks on the enemy landing force. On May 27th four Army heavy bombers and nine Navy fighters carried out a daylight attack against fierce air opposition, all but four fighters failing to return. Kuzume would need more than that to launch a determined attack that would succeed in pushing the enemy back into the sea. Consequently, on May 29th, General Numata and Admiral Senda had requested the immediate commitment of fleet and air strength into the Biak battle. They both relayed this message “The enemy apparently found the difficulty of rapid occupation of the airfield sector. The enemy will change, in all probability, its policy to occupy the whole island of Biak after the arrival of reinforcements, securing its present positions with landed units for a while. The officers and men on Biak Island are firm in their resolution to crush the enemy. However, our operations are severely restricted by the uncontested superiority of the enemy's feet and air units. The Biak Detachment, which is making every effort in destroying the confronting enemy, request for further support by the army and navy units concerned. We believe that the immediate commitment of our air forces and, if possible, some fleet units would give us a splendid opportunity to turn the tide of battle in the whole Pacific area in our favor.“ This finally convinced Admiral Toyoda to send reinforcements to the island.  To counter the Allied advance to Biak, the IJN dispatched from one third of its available naval land-based air strength from the Central Pacific to reinforce the 23rd Air Flotilla in western New Guinea. On May 28th 70 carrier-type fighters, 4 reconnaissance bombers, and 16 medium bombers were dispatched to western New Guinea. Another group of planes, comprising 48 fighters, 8 reconnaissance aircraft, and 20 bombers, were sent to western New Guinea and Halmahera from the Carolines on or about May 31st. On 29-30th May the flotilla carried out fresh attacks on the Biak landing force. On May 29th, sixteen medium bombers attacked the enemy fleet in the sea near BIAK Island before daybreak of that day, yet none of them returned. Furthermore, in a daylight attack on the same day, four Zero fighters strafed BIAK Island. None of them returned to the base either. On May 30th, the unit of the Zero fighters of the Navy again fired upon enemy ships in the sea off Mokmer. The damage on the enemy ships was not confirmed. However, the unit reported that they fought four P-38s and four B-25s of the enemy and shot down two B-25s above BIAK Island. Also as part of Operation KON, a huge task force under Admiral Sakonju, which included the battleship Fuso, four cruisers and eight destroyers, was to transport Major-General Tamada Yoshio's 2nd Amphibious Brigade towards Biak. Additionally, it was decided to move three infantry companies of the 35th Division from Sorong to Biak by barge. Sakonju's convoy finally left Davao on the night of June 2nd. In connection with KON Force's advance, the Japanese had planned heavy air strikes against Biak which were to be carried out by the recently reinforced 23rd Air Flotilla and the few army aircraft which remained at bases within range of Biak. Between 1645 and 1700 on 2 June, from eleven to fifteen Japanese planes bombed Allied positions on Biak, causing a few casualties and some light damage. Seven of these planes were shot down by shore-based anti-aircraft weapons, while guns aboard Seventh Fleet ships lying off Bosnek accounted for at least one more. Later during the same night, a few more enemy planes dropped some bombs harmlessly on and near Owi Island. Still more approached Biak during the night, causing many red alerts but not dropping any bombs. The next night, that of 3-4 June, no Japanese planes attacked Biak, although an unknown number bombed Owi Island without causing any damage or casualties. Again, however, enemy aircraft flew many reconnaissance flights around Biak, causing an almost continuous red alert until the early morning hours of 4 June. Early on the morning of June 3rd, at a point just east of the Talaud Islands, between Mindanao and Morotai, a 7th Fleet submarine sighted the Transport and 1st Screening Units and was in turn sighted by ships of the latter organization. Alongside this 7th Fleet PB4Y's, operating from Wakde Island, kept the Japanese vessels under surveillance the rest of the day, reporting that the course and speed of the enemy ships could bring them into range of Biak during the evening of June 4th. Their discovery by Allied aircraft so far from Biak apparently had not been anticipated by the Japanese, who later reported that they had not known Allied aircraft were capable of such long-range reconnaissance. Nevertheless, the three KON Force elements steamed on toward Biak, probably hoping that friendly aircraft might drive off the Allied reconnaissance planes and also protect the sea approaches to Biak. To further muddy the situation, Sakonju received false reports that a strong American carrier group was approaching the waters east of Biak. Admiral Kinkaid had indeed dispatched a special task force to deal with this threat, yet the warships could only arrive off Biak on the night of June 4th and didn't include any aircraft carrier. Nonetheless, knowing that he had been discovered and unwilling to risk so many ships under these circumstances, Sakonju would have to suspend the reinforcement run and turn back to Davao and Sorong.   When the Japanese called off KON on June 3rd, the Transport and the 1st and 2nd Screening Units were a little over 500 miles northwest of Biak and about 250 miles east-southeast of the Talaud Islands. At this point, the three forces were reorganized. The Transport Unit, accompanied by the three destroyers of the 1st Screening Unit, changed course for Sorong, while the 2nd Screening Unit and the two heavy cruisers of the 1st turned back toward Davao, which they probably reached late on June 5th. Of the ships moving to Sorong, the Fifth Air Force claimed to have sunk one destroyer and damaged at least two others. The Transport Unit and the 1st Screening Unit's three destroyers arrived safely at Sorong during the evening of June 4th. The Detached Unit, which had been moving toward Biak from Zamboanga on an independent course far to the west of the other three sections of KON Force, had also changed its direction during the night of 3-4 June, and reached Sorong sometime on the 4th. At Sorong the Transport Unit unloaded the 1,700 men of the 2nd Amphibious Brigade. The six destroyers of the Transport and 1st Screening Units then proceeded southwest to Ambon where they refueled. The Transport Unit's one heavy cruiser and one light cruiser sought shelter in Kaboei Bay, Waigeo Island, about 60 miles northwest of Sorong. On 6 June the heavy cruiser Aoba was attacked there by fifteen B-24's of the Fifth Air Force. First reports were that at least two hits were scored on the cruiser, but it was later learned that the ship suffered no damage. Instead, it was able to take part in a second KON Operation. Back over at Biak, Newman resumed the advance westward on the morning of June 3rd, making painfully slow progress because of the difficult terrain and lack of adequate supply lines. Meeting no opposition, they would finally dig in half a mile from the point at which the main ridge left the coast and turned inland near Mokmer. That day, however, Fuller learnt about the possible enemy naval attack, so he decided to halt any offensive actions for the moment. On June 4th, upon learning that no enemy carriers were in the Biak area, Sakonju was again ordered to prepare to run the American blockade, this time bringing the bulk of the 2nd and 3rd Battalions, 219th Regiment. There would be two naval groups, the first was the Transport Unit, containing three destroyers which had been part of the first KON Operation Transport Unit. The second section was the Screening Unit, also comprising three destroyers. For the second KON Operation there were two detached units, the 1st had one heavy and one light cruiser while the 2nd Detached Unit included the small craft and patrol boats which had put into Sorong at the end of the first KON. The three destroyers of the Transport Unit were each to embark 200 infantrymen at Sorong. In addition, the destroyers of either or both the Transport and Screening Units were each to tow to Biak one landing barge crammed with troops, probably 30 to 50 men to a barge. After two quiet nights, meanwhile, Newman decided to send three battalions forward toward the north-south section of the main ridge northwest of Mokmer on June 5th. Warned by the regimental commander that it was important to secure a foothold on the ridge before the Japanese could deny it to the 186th Infantry, the three assault battalions started westward about 8:00 on June 5th. Lack of water again slowed the advance. No water had been received in the forward area since the morning of the 4th, and Colonel Newman had ordered the troops westward against the advice of his staff and battalion commanders. About noon, however, a heavy rain fell. The regimental commander ordered all troops to halt, catch the rain in ponchos, and fill their canteens. "Had it not been for this lucky break, we would undoubtedly have had to halt in midafternoon." As events turned out, no Japanese opposition was encountered, and by 1500 the 3rd Battalion, 186th Infantry, was within 500 yards of the main ridge. Although Newman and General Doe then wanted to secure the dominating terrain north and northwest of the airfield, they would receive direct orders from Fuller to immediately seize Mokmer Drome and a beachhead on the coast directly south of that strip. Throughout the morning of June 6th the 186th Infantry directed most of its efforts to bringing supplies up to the forward units. Almost the entire 2nd Battalion was engaged in hand-carrying supplies to the 3rd Battalion atop the ridge, while the latter unit sent patrols toward Mokmer Drome seeking good routes of approach to that objective. About noon Colonel Newman reported to task force headquarters that no good route had been found and that supplies, especially the ever-needed water, had not been brought forward in sufficient quantities to allow a regimental attack to be launched that day, and he therefore recommended that the attack be postponed until June 7th. General Fuller approved this suggestion. The lack of supplies and water would delay the attack, however, though the 3rd Battalion would be able to move down the west side of the main ridge to take up positions along a line of departure for the next morning's attack. To support the infantry attack, on June 7th, a thirty-minute artillery concentration began at 7:00 that morning. The 146th, 205th, and 947th Field Artillery Battalions, from positions along the coast to the east, were registered on targets in the airfield area ready to support the advance, but most of the firing was undertaken by the 121st Field Artillery from its location behind the 186th Infantry. While the artillery fired on Mokmer Drome and along the low ridge between that field and the 186th Infantry, Fifth Air Force bombers attacked the Borokoe Drome area and also struck some targets along the low ridge. The airfield was only occupied by the 108th Airfield Construction Unit, which immediately fled the area because of the heavy bombardment. Newman's 1st and 3rd Battalions advanced south encountering no resistance as they crossed Mokmer Drome and reached the beach.  When, on 5 June, the 186th Infantry had reached the crest of the main coastal ridge, it had been on the left rear of the Japanese defenses on the low ridge and terraces above Mokmer Drome. Thus, the regiment had been in a favorable position to take these defenses from the rear. But in its move to the airfield, the 186th Infantry had bypassed the Biak Detachment's principal defensive positions. The bypassing had not been intentional. Colonel Newman had instructed both leading battalions to halt on the low ridge, reconnoiter along it in both directions, and report on Japanese defenses before moving on. According to Colonel Newman: "I received a negative report from both battalions, and ordered the movement to the airdrome. Evidently, the right battalion had failed in this patrolling effort." Instead, the 186th had captured its main objective, but now found itself surrounded by Kuzume's strongest defenses. The Japanese immediately began to pound the new American perimeter, with an artillery duel soon erupting. By nightfall, it had become impracticable to supply the 186th Regiment over the inland plateau road, which ended on the east side of the main ridge. From that point, all supplies would have to be hand-carried to Mokmer Drome and supply parties would be endangered by Japanese patrols, a few of which moved in behind the 186th as the regiment reached the beach, so the 3rd Battalion, 163rd Regiment would be dispatched to push over the inland plateau and protect the parties. Overwater supply was also attempted, yet as the first boats approached the shore they were greeted by machine gun and rifle fire from Japanese whom the 186th Infantry had not yet cleaned out of caves along the water line in front of Mokmer Drome. The small craft returned the fire, but were finally forced to withdraw. The 186th Infantry, according to Colonel Newman, was "glad to see them withdraw since they had our troops running for cover." At 2:00 another attempt was made to land supplies at Sboeria. The three LCM's managed to put their tanks ashore in the face of continuing Japanese fire, but accompanying LCT's were driven off by Japanese artillery. Two of the LCM's were so damaged by enemy fire that they could not fully retract their ramps and had to proceed the nine and a half miles back to Bosnek in reverse. Plans were made to effect all delivery of supplies and evacuation of casualties at night until the enemy fire on the Sboeria beachhead could be neutralized. The tanks which had been landed lumbered along the shore road fronting Mokmer Drome, destroying several small bunkers along the beach. Then they wheeled toward the low ridge north of the airfield, taking under fire a Japanese 75-mm. mountain gun and a 20-mm. piece which had opposed their landing. These two weapons were silenced. Moving cautiously northwestward from the field along a road which crossed the low ridge, the tanks destroyed two large pillboxes. Alongside this, Fuller sent two companies of Haney's 3rd Battalion to land on the Parai Jetty in order to outflank the Ibdi Pocket, which the 162nd had been unsuccessfully attempting to dislodge since the start of the month. But June 7th would also see the start of Operation KON's second attempt.  After rendezvousing off Misoöl Island that morning, Sakonju instructed his 8 destroyers to proceed to Biak. Air cover was to have been provided by planes of the 23rd Air Flotilla. But the cape area was being patrolled by Allied aircraft on June 8th and, about 1:30, the 23rd Air Flotilla cover of six planes was shot down or driven away by 5th Air Force P-38's.  Finding the air now free of enemy planes, American B-25's dived to the attack th convoy, reporting the convoy as 2 light cruisers and 4 destroyers. Initially, it was claimed that 1 destroyer was sunk, 2 were left sinking, and the fourth was damaged. A few days later, destruction was reassessed as 4 destroyers sunk and 2 light cruisers chased to the northwest. These claims were exaggerated. One destroyer, the Harusame, was holed by a near miss and sank rapidly, the bulk of its crew being saved. Another destroyer was damaged by a bomb and took some water; two others were slightly damaged by strafing. Neither speed nor navigation was impeded for any of the three. The two light cruisers reported by the Allied planes were, of course, the other two destroyers. These two might have taken some evasive action by heading northwest for a short time, but as soon as the Harusame crew had been rescued and the Allied planes had disappeared, the convoy reformed and continued on toward Biak. The convoy reformed and continued on toward Biak, undeterred by reports of strong enemy elements in the area. By nightfall, however, it was on a collision course with the cruisers of Admirals Crutchley and Berkey.  At about 6:00 on the 8th, the Transport and Screening Units received a report from a Japanese aircraft that an Allied naval force comprising 1 battleship, 4 cruisers, and 8 destroyers was moving west at high speed from an undesignated point east of Biak. This report was at least partially correct. The Allied task force which had been formed on June 3rd had again assembled on the 8th, having been alerted by reports of the air-sea battle off the Kaap de Goede Hoop. But the Japanese convoy commander apparently took this air reconnaissance report with at least one grain of salt--had not similar information received on June 3rd proved inaccurate? The Transport and Screening Units steamed on, despite the fact that the Kaap de Goede Hoop action had put the force behind schedule. At 11:30 the two enemy groups were approximately forty miles off the north coast of Soepiori Island, ready to turn southeast toward Korim Bay, on the northeast side of Biak. Minutes later a destroyer in the van sighted the Allied task force heading northwest around Biak. The convoy commander quickly realized that he was badly outnumbered and decided that discretion was called for. The Japanese convoy slipping towards the Mapia islands, seeing the allied destroyers failing to pursue them. Yet that is it for Biak for now as we now need to head over to the Wakde-Sarmi front.   General Sibert was preparing to resume the westward offensive. By June 14th, the 20th Regiment had relieved the 158th at the Tirfoam River; and although Sibert wanted to complete unloading of his remaining units before sending the 20th to push westward, General Krueger ordered him to start an immediate offensive on June 18th. Now, however, they were up against almost the full strength of General Tagami's 36th Division. Company B pushed on toward the village at the entrance to the defile between Lone Tree Hill and the eastern nose of Mt. Saksin. This advance was greeted by a hail of fire from Japanese automatic weapons emplaced in the defile--fire reminiscent of the opposition encountered by Company B, 158th Infantry, at the same place more than three weeks earlier. The 20th Infantry's Company B tried to outflank the enemy position to the south but was halted by intense Japanese machine gun fire. Tanks sent forward to aid the infantry were unable to reach the enemy guns because the terrain was impassable to tracked or wheeled vehicles, which could scarcely negotiate the rough road, let alone the thick jungle and rising ground to the south. Late in the afternoon Company A was sent forward to Company B's position, but both units encountered heavy fire and soon lost contact with the rest of the 1st Battalion. The two companies remained for the night in an isolated perimeter near the village and about 400 yards west of the main body. The 3rd Battalion had moved north off the coastal road during the morning, and late in the afternoon it had established a perimeter extending south 200 yards from the beach along the east bank of the Snaky River. The battalion had encountered little opposition during the day, but patrols which had crossed the Snaky before dark reported finding many Japanese defensive positions on the eastern slopes of Lone Tree Hill. A gap which existed between the 1st and 3rd Battalions was partially filled just before nightfall by elements of the 2nd Battalion, which were sent forward late in the afternoon. Casualties during the day were four killed and twenty-eight wounded. The 1st and 3rd Battalions, 1st Infantry, moved across the Tor River in the morning of June 20th and took over the positions in the vicinity of Maffin No. 1 vacated by the 20th Infantry. The 2nd Battalion, 1st Infantry, assumed responsibility for the protection of the bridgehead across the Tor.  The following morning, Sibert then directed his units to patrol extensively in order to locate enemy strong points on and around the hill. Thanks to the information gathered, the 3rd Battalion would attack towards Rocky Point in the afternoon. At the top of Lone Tree Hill was a stretch of rough but generally level ground lying mostly along the western part of the hill. This flat ground, about 700 yards long north to south, was shaped like a crude dumbbell. At its northern end, the level area was about 300 yards wide. It narrowed at the center of the hill to less than 100 yards but broadened again on the south to a width of about 250 yards. There were many coral outcroppings, potholes, and small crevices, while on the north the hill terminated in a very rugged prominence called Rocky Point. This terrain feature, which extended into Maffin Bay from the central mass of Lone Tree Hill, was about 300 yards wide east to west. Its northern face was not as heavily overgrown as the rest of Lone Tree Hill. Although Rocky Point's northeast slope was steep, foot troops could climb that face with more ease than they could approach the top of Lone Tree Hill from most other points. A deep ravine ran southwest into the central mass of Lone Tree Hill from a sandy beach on the east side of Rocky Point. The floor of the ravine varied from 20 to 30 yards in width and its nearly vertical western wall was 40 to 50 feet high. Both sides were honeycombed with natural or man-made tunnels, caverns, and small caves, most of which were connected with each other by underground or deeply defiladed passages. Some caves reached a width of 40 feet, a depth into the hillside of 50 feet, and a height of 20 feet. The ravine terminated on the eastern slope of Lone Tree Hill in a steep grade at the narrow central portion of the hilltop. At 1:45pm, after a fifteen-minute artillery and 4.2-inch mortar preparation, one company moved across the Snaky River, immediately finding the twenty-foot cliff along the eastern side of the shelf which lay between the Snaky River and the central mass of Lone Tree Hill. The morning patrols had not, apparently, reported the existence of this cliff, and naturally it was not known that Japanese defenses were established along it. Machine gun and rifle fire from the 1st Battalion, 224th Infantry, soon pinned down the 3rd Battalion's leading platoon. The company commander quickly sent part of his unit northward to find the Japanese left flank. Moving around the northeast end of the shelf, this group discovered the beach entrance to the deep ravine between the western side of the shelf and Rocky Point. Progress into or across the ravine was impossible in the face of the intense Japanese small arms fire which greeted the advancing American unit. Company B, 6th Engineers, then in the forward area to cut a road from the mouth of the Snaky River to Rocky Point, was brought up to the ravine to help clean out caves and crevices with flame throwers and demolitions, but could not reach the enemy positions through the continued machine gun, mortar, and rifle fire. Infantry bazooka squads also tried to blast the Japanese out of their caves but failed when their ammunition ran out. Since there was no time to bring additional rockets forward before dark, all elements of the 3rd Battalion and the engineer company were withdrawn to the east bank of the Snaky River for the night. The 20th Infantry was to continue the assault on the morrow with the 3rd Battalion moving against Lone Tree Hill from the northeast, the 2nd Battalion in reserve, and the 1st Battalion remaining in its holding position. On the morning of June 22, after a heavy air and artillery concentration on Rocky Point, the 3rd Battalion once again attacked northwest with Companies K and I, successfully driving the Japanese back into their caves to reach the top of Lone Tree Hill just south of Rocky Point. Meanwhile, another two companies had attacked southwest to force their way up the southeast slope of the hill; but subjected to heavy machine-gun fire, they would have to withdraw and march north to join Companies K and I. The 2nd Battalion also moved forward and took positions to the south of the 3rd Battalion. Worried about the American gains, Colonel Matsuyama personally led two companies in the afternoon to fall on the 3rd Battalion's perimeter with suicidal fury. Confused fighting, sometimes hand-to-hand, continued well into the night, with Matsuyama himself getting shot on the thigh. Yet this attack would successfully position the Japanese companies on the rear of the 2nd and 3rd Battalions, thus cutting them off from regimental headquarters. Matsuyama also recalled his 2nd Battalion from the Maffin area, so he would employ these reinforcements to attack Sibert's 2nd Battalion on June 23. At dawn on the 23rd Japanese troops, some of whom were using American weapons and wearing parts of American uniforms, attacked the 2nd Battalion, 20th Infantry, from the deep ravine. The battalion initially held its fire, thinking that the enemy force might be a friendly patrol, and the Japanese were able to advance to within fifteen yards of the battalion lines before being recognized. It was an hour before the results of this error could be corrected--an hour during which both the 2nd Battalion and the Japanese suffered heavy losses. The hour ended with an enemy retreat. Isolated, the 2nd Battalion then decided to withdraw and march north towards the 3rd Battalion's perimeter at the top of the hill, getting harassed all the way by Matsuyama's forces. During the night, the Japanese launched a banzai charge against the perimeter, getting very close to retaking Lone Tree Hill but suffering heavy casualties in the end. Upon learning that his battalions were cut off, meanwhile, Sibert decided to outflank the hill by a shore-to-shore maneuver and then continue the attack from both west and east. Accordingly, Companies K and I of the 1st Regiment boarded ten LVTs on the morning of June 24th and moved to the beach just west of Rocky Point, under the protection of the 6th Reconnaissance Troop. Both companies would land successfully by midday against strong Japanese fire, though they would be rapidly pinned down on the narrow beach. Thankfully, Sibert also landed four tanks two hours later to secure the beachhead. This diversion would allow the 2nd and 3rd Battalions, 20th Regiment to begin clearing the Japanese from the many caves and crevices on Rocky Point, the deep ravine east of the point, and the hilltop plateau, further securing the supply route up the hill.  By nightfall, no enemy counterattack developed, as Tagami had instead decided to withdraw the 224th Regiment to the Hill 255-Mount Saksin line while the 223rd Regiment retreated behind the Woske River. Thus Matsuyama's resistance in the area weakened and the Americans were finally able to clear Rocky Point. The next day the 2nd and 3rd Battalions, 20th Infantry, continued clearing Rocky Point, the deep ravine, the northern part of the hilltop plateau, and the eastern shelf, where a few scattered Japanese still held positions along the twenty-foot-high cliff. Flame throwers, demolition charges, bazookas, and hand grenades all proved successful in eliminating Japanese resistance and sealing or clearing caves and crevices. The task was easier on the 25th, for the Japanese slowly gave up the fight and were killed or sealed off in their caves. Casualties continued to mount, the 2nd Battalion, 20th Infantry, had only about two hundred effectives by the end of the day but many of the losses were not due to Japanese action. Many men were evacuated over the now secured supply route to the top of the hill as they fell from exhaustion or became sick. On the beach west of Rocky Point Companies I and K, 1st Infantry, had little success in expanding their beachhead. The tanks proved useless in the area and were therefore withdrawn to Maffin No. 1. The two infantry companies, pinned down during the morning, kept up a continuous mortar barrage against Japanese positions in the swamp to the south, against the western cliff of Lone Tree Hill, and, when certain such fire would not endanger troops atop the hill, against the northwest corner of Rocky Point. This mortar fire, coupled with the operations on the plateau, began to have the desired effect during the afternoon, and Companies I and K were able to push their defenses beyond the narrow beachhead slightly southward and westward and toward the shore beneath Rocky Point. Once or twice during the afternoon, patrols were able to reach the top of Lone Tree Hill from the northwest corner of the point and established contact with 20th Infantry units. Late in the afternoon Company M, 1st Infantry, operating from the east side of the point, managed to push a patrol around the shore to establish contact with Company K. Though Companies I and K could find little tangible evidence of the results of their operations, they had actually wiped out the 223rd Infantry's defense force in the area just west of Lone Tree Hill. By dusk on the 25th, it had become obvious that the combined efforts of the 3rd Battalion, 1st Infantry, and the 2nd and 3rd Battalions, 20th Infantry, had either cleared out the northern half of Lone Tree Hill or had forced the Japanese to withdraw. The latter conclusion was the more nearly correct. The 36th Division decided on 25 June to withdraw the bulk of the Center and Right Sector Forces west of the Woske River and establish new defensive positions, thereby keeping the 223rd Infantry, the bulk of which had not been committed to action in the Lone Tree Hill area, more or less intact. Only the remnants of the 224th Infantry were to remain east of the Woske, and they were to withdraw into rough terrain southwest of Mt. Saksin. At nightfall on the 25th, General Sibert estimated that his three forward battalions had lost approximately 140 men killed and 850 wounded and evacuated, including those who had to be sent back to the rear because of wounds, sickness, heat exhaustion, or psychoneurotic disorders. Known Japanese dead in the northern part of the hill numbered 344, but it could not be estimated how many more had been thrown over the west cliff, sealed in caves, or carried off by withdrawing remnants of the Japanese defense force. According to Japanese sources, the Japanese had lost about 500 men killed and another 300 wounded in the Lone Tree Hill-Hill 225-Mt. Saksin area.   I would like to take this time to remind you all that this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Please go subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry after that, give my personal channel a look over at The Pacific War Channel at Youtube, it would mean a lot to me. The Chinese were accredited with the fall of Moguang despite the Chindits taking the lionshare of the fighting. Things were advancing very well for the allies in the new Biak front. As for the battle for Lone Tree Hill, it was a costly one, and not one the Americans or Japanese would soon forget.   

Last time we spoke about the battle of Wakde. Operation Tornado, the amphibious assault of Wakde island faced many logistical challenges, but not as much Japanese resistance. The landing at Arare was a large success, seeing the allies secure a beachhead before advancing inland. Through a combination of air, naval and ground attacks, the smaller Japanese force was pushed back. Likewise the islands of Liki and Niroemoar were taken with relative ease as well. Over in the Burma-India theater, the battle for Kohima was gradually seeing the Japanese fail to meet their objectives. Without ammunition or provision supplies flowing to them, the Japanese at Kohima had no hope of holding back the allies. Thus the Japanese were beginning to fall back and now were setting up a defense to stop the allies from entering central Burma. Meanwhile a siege was being erected against Myitkyina. This episode is the Siege of Myitkyina  Welcome to the Pacific War Podcast Week by Week, I am your dutiful host Craig Watson. But, before we start I want to also remind you this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Perhaps you want to learn more about world war two? Kings and Generals have an assortment of episodes on world war two and much more  so go give them a look over on Youtube. So please subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry for some more history related content, over on my channel, the Pacific War Channel you can find a few videos all the way from the Opium Wars of the 1800's until the end of the Pacific War in 1945.  Last time we saw General Stilwell had managed to seize the Myitkyina airstrip. Now allied forces were gradually securing northern Burma, pushing further towards Mogaung. The Japanese 18th Division were absolutely battered and barely holding on around Kamaing. However when the American-Chinese forces seized Myitkyina's airstrip, they had failed to quickly attack the main town, which the Japanese were reinforcing heavily. Now they were being attacked from both ends. This prompted General Stilwell to dispatch General Boatner with the task force to try and turn things around. Simultaneously, Stilwell had just received word of the Chindits abandonment of the Blackpool stronghold, outraged by this he ordered them to advance northwards to support the attack on Mogaung. Three brigades, the 111th, 14th and the 3rd West Africans were to advance north to support Calvert's 77th Brigade to take Mogaung. The Chindits were incredulous. Nonetheless, Lentaigne, pressed by Stilwell, sought to have his 3rd West Africans, the 14th and 111th Brigades squeeze the Japanese along the western flanks of the Mogaung valley, in which Mogaung occupied the lowest point of an inverted triangle, with the other two points being Kamaing on the top left and Myitkyina on the top right. With any luck, his three brigades could capture Kamaing, which occupied an enviable place on the Mogaung River. The 111th Brigade was to move north-east and operate in the Pahok-Sahmaw area, destroying dumps and blocking enemy movement. Thebattered men of the 111th Brigade struggled to cope with the march north, to the hills east of Lakhren and west of Mogaung. Conditions were appalling on the three-day march to Lakhren village. From there, the best approach to Mogaung would be identified. They were making for a waterlogged, mosquito-infested area north east of the Lake. This area was also alive with Japanese units. They were required to support 77th Brigade's attack on Mogaung, by pushing from the west as Calvert's command advanced from the south-east. However, 111th Brigade was spent after Blackpool and many men were forced to return to Lakhren village. There was also a new task for 14th Brigade – the capture of Taungni. With the casualties evacuated, 14th Brigade abandoned the Kyunsalai Pass and headed north towards Mogaung. The steamy swamps continued to take their toll within the ranks of the York and Lancaster columns. Dysentery was rife. Animals collapsed and were shot where they dropped. They reached Mokso on June 25. This much-used rallying point, consisting of four huts, was a sea of mud and detritus, decorated with black clouds of flies. Rations were being consumed rapidly in an area devoid of drop zones. Brigadier Brodie, 14th Brigade's Commander, responded to the extreme circumstances. His men, despite their dangerously poor condition, were still expected to harass the Japanese along the railway and support 77th Brigade's assault on Mogaung. He formed his columns into “Light Battle Groups”, free of heavy weapons and the wounded and sick, now lying together in the mud. Meanwhile, some Nigerian units were now struggling on the road to Lakhren along the reailway, as were the York and Lancaster sick – around 300 in all, 200 of whom were dangerously ill. On May 25th, the same day the Chindits had quit Blackpool, Stilwell ordered the Morris Force to seize Waingmaw, across the river from Myitkyina. Unfortunately, the Japanese had entrenched themselves at the town and enjoyed the luxury of a natural moat after heavy rains flooded the fields on the approaches to the town. Morrisforce was not a proper brigade, having only two columns of troops, to which a third from the 111th Brigade had been added 1,500 troops in total. It had been conceived to harass the Japanese on jungle areas. Now, they were up against fortified positions. The result was a bloodbath. Morrisforce then began to rapidly deplete in strength. By July 14th, it was to consist of exactly three platoons, roughly 120 men. On May 31st, Boatner then launched his first coordinated attack against Myitkyina seeing the 42nd regiment reach the Waingmaw ferry road. Built up twelve feet above the neighboring paddy fields, the road gave the Japanese a magnificent defensive position, which they exploited cleverly. The Chinese recoiled from this natural fortification but were able to beat off a Japanese counterattack. The 150th Regiment reached the riverbank and drew up in an arc about a sawmill in which the Japanese had a strongpoint. Meanwhile Colonel Hunter's 2nd battalion reinforced with engineers advanced to Radhapur where they were heavily counterattacked by the 114th Regiment. The next day, the inexperienced 236th Engineer Battalion was sent against Namkwi. The motive behind the 236th's attack was to contain the Japanese in the Namkwi area and introduce the battalion to combat under relatively easy conditions. One company of the 236th did succeed in entering Namkwi but instead of promptly consolidating to meet the inevitable Japanese counterattack fell out for a break. The Japanese counterattacked and drove the unwary engineers right back out of the village. Both the engineers and the 2600 replacements of the Galahad Unit that had recently arrived lacked adequate experience fighting the Japanese and as such suffered badly when fighting against them. Colonel Hunter's veteran Marauders, however, had suffered ample casualties and thus needed these green replacements in order to continue existing as a fighting force. Boatner tossed a last ditch effort on June 3rd, but his Chinese forces had suffered 320 casualties and their ammunition was running low. While he waited for supplies to build up, he used his green american troops to give them some experience, the Chinese meanwhile tunneled towards the Japanese still suffering heavy casualties. This allowed more Japanese troops to break through and reinforce Myitkyina, with a huge relief force of the 52rd Division soon on its way. Meanwhile, Lieutenant-General Matsuyama Yuzo of the depleted 56th Division was facing a deadly offensive of his own on the Yunnan front. After a series of negotiations between Chiang Kai-Shek and the Americans, it had been agreed, General Wei Lihuang's Y Force would cross the Salween River to attack the Tengchong and Longling areas. This would effectively allow the allies to link the Ledo Road with China, bypassing the heavy Japanese concentration along the Burma Road. Although the Americans had supplied the Y Force with artillery and ammunition, the Chinese had failed to bring the Y Force divisions up to strength, and many questioned their training. Regardless, General Wei planned to have elements of the 20th army group cross the Salween at several points before initiating a full assault against Tengchong once reinforcements had been ferried through. Further south, elements of the 11th Army Group would also cross the Salween to launch attacks on Pingda and Longling.  On the night of May 11th, the Chinese forces began to cross the Salween River against little Japanese resistance, since Matsuyama had decided not to defend the crossing sites, instead placing his main line of resistance along the ridge line some 10 miles west. The 198th Division was able to assemble in front of the Mamien Pass; the 36th division was successfully ferried through Mengka; the brand new 39th Division managed to secure the Hueijen bridge and the 76th and 88th Divisions were converging on Pingda for the attack. On May 12th, the 198th were making good progress against the Mamien Pass as the 36th had surrounded the Japanese outposts in the eastern end of the Tiantouzhai pass. However Colonel Kurashige Yasuyoshi had his 148th regiment launched two surprise night counterattacks. All that night, the Japanese quietly filtered down from a nearby ridge and assembled near the Chinese position. Attacking at dawn, they surprised the Chinese and almost wiped them out before aid came. When darkness came, the 36th were at their bivouac. That night, the Japanese attacked vigorously, overrunning the division command post and causing the flustered 36th Division to fall back to the Salween. At dawn, the 53rd Army commander, Maj. Gen. Chou Fu-cheng, pushed a regiment across the Salween and restored the situation by attacking the Japanese flank. General Chou was an aggressive and tenacious fighter, whom his Manchurian soldiers had nicknamed Old Board-Back, and who had the reputation of never having yielded an inch to the Japanese. But even Chou could not immediately restore the morale of the 36th Division, which for some weeks took no further part in the Ta-tang-tzu fighting, and the rest of the 53rd Army had to bear the burden of clearing the pass. In response to this, Lt General Zhou Fucheng of the 53rd Army pushed the 116th and 130th Divisions across the river to resume the advance towards Tiantouzhai. At the Mamien pass, the 592nd and 594th Regiments began clearing out the Japanese strongpoints while the 593rd Regiment moved west over mountain byways to emerge into the Shweli valley by the 16th, securing the western end of the pass and forcing Kurashige's men to withdraw into the fortified village of Shangzhaigongfang. To the south, the New 39th would be able to secure Hemushu by the 17th. Yet the Japanese under Colonel Matsui Hideji soon recovered from the initial surprise of the Chinese offensive and the 1st battalion, 113th regiment drove the Chinese from Hung-mu-shu. The Japanese exploited their success and pushed the entire New 39th Division back against the Salween. Further south, the 76th Division met outposts of the 1st battalion, 146th regiment and forced them back to the heights overlooking Ping-ka. Meanwhile the 88th Division from the north was fighting through a series of fortified villages as it headed south to join the 76th Division. By the 16th, thirteen villages were occupied in the area northeast of Ping-ka, but the Japanese, as they withdrew, received reinforcements. Strengthened by the 2nd battalion, 113th regiment, the Japanese did not attack the Chinese who were pursuing them, but moved south and hit the Chinese 228th Regiment south and west of Ping-ka.  Sensing the imminent danger from the north,  Matsuyama redirected the 2nd and 3rd Battalions of the 113th Regiment to reinforce Shangzhaigongfang, where Kurashige would continue to resist staunchly. Reports from the American liaison teams were not cheerful. Americans observing the Kaolikung Range actions found that Japanese fire was accurate and economical, and that the enemy's use of camouflage and concealment approached perfection. The Japanese revealed no disposition to surrender though they were heavily outnumbered, often surrounded, and had neither air support nor air supply. On the other hand, the Americans reported that the Chinese endlessly wasted manpower and ammunition in costly frontal attacks. They reported that relations with the Chinese were not always as friendly as had been hoped, and they believed there would have to be better cooperation between Chinese and Americans if the Japanese were to be defeated. The Chinese were described as merely tolerating the Americans' presence and as paying little attention to their advice. The liaison personnel freely admitted their own shortcomings, and by their reports suggested that patience was the most important quality for a liaison officer to cultivate when dealing with the Chinese. Matsuyama ha spread his forces widely, over a near 60 mile sector, thus he would be unable to perform mobile operations and was gradually shifting to a defense of the Kaolikung mountain range by the 20th. This allowed the 53rd Army to continue their advance and the New 39th to retake Hemushu. By late May, with the 198th Division apparently contained on the north, Kurashige then rushed with the 113th Regiment to reinforce the Tiantouzhai front, where they successfully stopped the 53rd Army on its tracks. Yet on June 1st, the 54th Army, emerged in the Shweli valley from Ta-tang-tzu pass to join the 593rd Regiment from the Ma-mien pass. Next day the Chinese occupied Chieh-tou village and began patrolling the Shweli valley. They took the advice of Y Force officers and donned Burmese clothes. So disguised, they found it easy to enter Japanese-held villages. When taken by surprise, the Japanese were willing to abandon many of their outer defenses. With TNT charges dropped by the 27th Troop Carrier's C-47's when the weather cleared, the 198th Division also blew up the last pillboxes at Chai-kung-tang on June 13th. When the last shots had been fired and the Chinese farmer boys of the 54th Army reported the area secure, there was bewilderment at finding only 75 Japanese bodies in defenses that must have been manned by at least 300 men, and shock and nausea when the Japanese kitchens revealed how the defenders had been able to prolong their stay. Pitiful and ghastly evidence showed that the Japanese had resorted to cannibalism when their rations failed. On June 14th the Japanese further quit Chiao-tou-chieh, leaving many stores to the 2nd Reserve and 36th Divisions. In the end, the Kurashige Detachment would pull back to Watien while the Inose Battalion retreated towards Kutungchieh. General Wei then ordered the 71st Army to cross the Salween, just below the Huitung Bridge to attack Longling while a containing force attacked the strong Japanese harrison at Lameng. 20,000 troops of the 71st would cross the river by June 5th. At this time the 76th division left a regiment to besiege Pinga while the bulk pushed on to attack Matsuyama's HQ at Mangshi, while the 9th Division crossed the Salween, cutting off Pingda's line of communication. On June 4th, the new 28th Division attacked Lameng and forced Major Kanemitsu Keijirous garrison to retreat into the Mount Song fortress where they would hold out for several months. While the New 28th held down Kanemitsu, the 87th Division continued towards Longling, joining up with the 88th on June 8th to begin a siege of the main Japanese position on the Yunnan front. The Japanese still held onto the Burma Road east of Lung-ling, but as of June 7th the 87th Division had covered about two thirds of the distance along the Burma Road from the river to Lung-ling. The Chinese had to deal with small Japanese tanks, which had some moral effect but failed to halt the Chinese advance. Indeed, the latter had been fairly swift, for the Chinese had surprised the Japanese, and had been able to ease their supply problems by the capture of some Japanese rice depots.  To meet the new crisis, Matsuyama dispatched the 1st battalion, 119th regiment to stop the 2nd army at Xiangdaxiang and then attack Longling from the south. Matsuyama ordered Colonel Matsui Hideji to immediately relieve Longling from the north; and for the Yamazaki Composite Group to keep the Mangshi-Longling road open. Meanwhile, Lieutenant-General Song Xilian, had been making some progress against Longling's two outer heights, the stout defenders would manage to repel the uncoordinated Chinese assaults for about a week. Four days of un-coordinated infantry attacks, with little artillery support, failed to carry the three mountains inside Lung-ling, and there was nothing to show for the heavy drain on the 71st Army's ammunition stocks. This gave more time for Matsuyama's reinforcements to arrive by June 14th. On the 15th Matsui launched a heavy attack, in coordination with the besieged defenders, successfully defeating the 71st Army and consequently driving the Chinese from the vicinity of Longling. The Japanese had thus been able to drive back 10000 Chinese effectives by an attack with only 1500.  Repeated attempts by American liaison personnel with the 71st Army to learn how a small Japanese garrison could drive back a Chinese army group only brought embarrassed smiles from Chinese officers. The Chinese finally related on June 25th that the 261st Regiment had bolted, and that the commanding general of the 87th Division had attempted suicide. When fuller details of the fighting around Longling were available, anger and annoyance spread from Y Force to the Generalissimo himself. Y Force personnel considered the Chinese decision to withdraw from Longling inexcusable because the 11th Army Group had sent forward no reinforcements to meet the initial Japanese counterattacks. Of 21 battalions in the Longling vicinity on June 14th, only 9 took part in the fighting. In describing the defensive attitude of the 259th Regiment, as an example of the conduct that had cost the chance of a speedy breakthrough into Burma, one American liaison officer wrote: "From the time that we crossed the river until we reached Longling, the regimental commander continually had his troops in the rear digging emplacements and trenches in the fear that they would have to retreat." Yet that is it for the Yunnan battle for now as we need to head over to the Kamaing area. General Tanaka's 18th Division were withdrawing with the 22nd and 38th division in hot pursuit. To the south the Seton Roadblock was being held by 112th Regiment, threatening to thwart Tanaka's plans. Coming to their aid, General Sun sent his 113th and 114th Regiments to descend upon Lawa, where General Aida began an unauthorized retreat on June 4th. The abandonment of Lawa severely affected the defense of Kamaing, thus General Aida was relieved of command and replaced with Colonel Imaoka Soshiro. Colonel Shoshiro immediately began to dig in at Lagawng. Meanwhile the bulk of the 55th and 56th regiments were holding onto the Nanyaseik area as General Liao's 65th regiment cut their withdrawal route off on June 1st. While the rest of the 22nd Division were applying pressure from the north, the heavily outnumbered Japanese had no choice but to quickly cut a trail southeast, finally withdrawing on june 7th. Their retreat was chaotic, both regiments lost contact with each other and with their subordinate units as they made separate ways towards Kamaing. It would only be the 3rd battalion, 56th regiment who would arrive at Kamaing by June 10th, most of the others would reach Lakatkawng in late june. While retreating, their artillery units were intercepted by the pursuing Chinese and the artillerymen chose to die with their guns. While the 22nd Division and 149th Regiment thus pushed towards Kamaing, General Sun had also sent the 113th Regiment to take Zigyun and the 114th Regiment to advance southwards and support the Chindit assault on Mogaung. The Chinese were able to encircle Kamaing from the west, north and east as the 149th Regiment then moving across the fields and into Kamaing to take the settlement on June 16th,  pushing Tanaka's battle-weary troops to the hills south and west of Kamaing. Meanwhile Brigadier Calbert was able to reach the Mogaung area by the start of June. 12 miles from Mogaung, Calvert's forward troops began to run into Japanese patrols and snipers. Despite taking losses, the 3/6th Gurkhas led by Colonel Claude Rome, who in his previous incarnation had been overlord of “Broadway,” pushed on, seizing the western heights overlooking Mogaung on May 31. On June 1st, Calvert's South Staffords linked up with Rome and the Gurkhas at the village of Loihinche. Other elements of the brigade reached the southern foothills of the heights, three miles south of the town, on June 2 and went straight into the fray. That same day, the Lancashire Fusiliers and the South Staffords also attacked Lakum (held by some ordnance troops and a field hospital). The leading force of Fusiliers was soon pinned down by heavy fire, the impasse only broken when a Bren gunner in the leading rifle section went wild, and ran “straight up the hill, firing from the hip and screaming curses at the Japanese.” Softening up the Japanese with airstrikes from Air Commando Mustangs, troops of the 1st Lancashire Fusiliers and 1st South Staffordshires attacked and wrested Lakum village away from Japanese and occupied the summit of the range of hills overlooking the city, to the northwest.  Calvert decided he would need to build a base akin to White City, where he could collect supplies and build an airstrip to take out the wounded. Calvert fixed his eye on the village of Lakum, occupying a strategic spot on the eastern foothills of the heights overlooking the Mogaung plain. Upon capturing the village, the Chindits would also find substantial ammunition, a field hospital and buildings which had obviously hosted several Japanese headquarters. The country leading up to Lakum, however, was hard stuff. It was in the midst of thick jungle intersected by deep ravines. The path proved difficult to follow as it sometimes wound along a ridge and sometimes went straight up or straight down. The place was a defender's paradise. “A handful of resolute men could hold successive hill-tops for hours against a large force such as ours overburdened with mules and heavy stores,”. In response to the new threat, Colonel Okada Hakuji rushed over with some units of his 128th Regiment to protect Mogaung, leaving his 1st Battalion to face the rapidly-approaching 114th Regiment and immediately recalling his 3rd Battalion from Seton. Alongside this General Honda ordered General Takeda to turn back from Myitkyina and instead secure the Moguang-Kameing area. Thus the chance to lift the siege of Myitkyina was lost. By June 3rd, the Chindits had built a new airstrip near the Tapaw Ferry, allowing airdrops of supplies and equipment to spill in. Calvert was now ready to launch his attack.  Early on June 8th, the 1st South Staffords set off to secure the Pinhmi. The village was defended by elements of 3rd battalion, 128th regiment who were also protecting some ammunition dumps in the area. The Staffords routed the Japanese and destroyed the dumps, clearing the way to the bridge. By now it was afternoon, and they stepped aside to let the 1st Lancashire Fusiliers move on to capture the 150-foot-long bridge. All seemed well until a hail of gunfire shattered the silence, tearing into the Fusiliers. Two companies were pinned down in the ditch while another was in the jungle further down. At about 6 pm, Calvert arrived at the front to confer with Major David Monteith of the attacking company. It was decided that two platoons, under the cover of a mortar barrage would dash across the bridge and secure the other bank. Calvert's mind went to the 4.2” mortars. He intended to use them to deadly effect now. Two platoons of Fusiliers moved in on the bridge, with one platoon attempting to move along the ditch, only to struggle against the thick, waist-high grass, as the second slipped off the road into the jungle. The mortars, about 60 yards behind them, began firing, slowly at first, but then as fast as the men could drop bombs into the tubes. The men, with fixed bayonets, charged across the bridge. The Japanese waited until the British were halfway across before opening fire. Tracers filled that little space, bullets tearing into flesh. The Chindits toppled left and right. Some crawled in their bellies, trying to get just far enough to throw their grenades upon the enemy positions. By 6.15, it was all over, the retreat was called. Calvert summoned US airstrikes. Mustangs swooped in, bombing and machine-gunning the enemy emplacements, but one Mustang mistakenly bombed a group of Fusiliers unloading mules. Because of the unexpected opposition, Calvert would instead direct his forces to ford the Wettauk Chaung and take Mahaung and Ywathitgale, which successfully fell on June 9. The following morning, a Gurkha company attacked towards Kyaingyi and the railway to create the impression that the main attack would come from this direction while the bulk of the 3/6th Gurkhas made a wide right flank advance to attack the bridge from the rear. At dawn on the 10th, Shaw's Gurkhas moved forward, waist deep in marsh water and mud. The bridge assault party was under the command of Captain Michael Allmand, a one-time cavalryman commissioned into the Indian Army in 1942 after escaping from Singapore. Allmand moved his men forward warily. The approaches to the bridge were narrow with the road up on a high embankment with swampy, tree-heavy low-ground flanking both sides. Coming in from the marsh, the Gurkhas set upon the bunkers at the bridge with grenades and small-arms fire, but the Japanese held their ground. At 10 am, they tried again, shooting and hurling grenades from amid waist-deep mud of the Wettauk. Allmand, who was close to shore charged. Throwing grenades to scatter the enemy, he closed in to kill three with his kukri. Rallied by his heroism, the Gurkhas rushed the remaining defenders, capturing the bridge. About 35 Japanese were killed at the bridge and the Gurkhas captured one medium machinegun and two light machineguns. In return, Chindit casualties in the encirclement and capture of the Pinhmi Bridge came to about 130 killed and wounded. By mid-afternoon, Calvert had deployed two battalions up on the Mogaung-Pinhmi Road, while a third occupied the bridge area. Conquering the brigade saw the Chindits suffer 130 casualties, while Okada's troops had also suffered heavy casualties, yet they had successfully delayed the Chindit advance for four days, which allowed Takeda to bring the bulk of his forces back into Mogaung. Knowing full well that the Japanese had been able to reinforce Mogaung, Calvert decided to wait for the arrival of the Chinese troops that Stilwell had sent to reinforce him; but in the meantime, he launched a successful attack against the courthouse on June 11, followed by a failed advance towards the river. To secure his right flank, Calvert had decided to send a company of South Staffords under a new replacement officer, Major “Archie” Wavell Jr, son of the venerable Viceroy, to secure the area between the road and the Mogaung River. The Staffords made good headway, but near the river, they came under heavy fire from entrenched Japanese. Wavell was hit in the wrist, the bullet almost taking the hand off; and he was then pulled out of the line and walked back to the field hospital near Gurkha village, clutching onto his hand, which was now hanging on by a sinew. That night, the surgeons at Gurkha Village were to remove the hand entirely. Wavell Jr's war was over. In the end, the Staffords also had to pull back towards the road after the loss of their commanding officer. By June 15th,  the Chinese had still not appeared, thus the Chindits ultimately had to withdraw to Pinhmi.  Back over at Myitkyina artillery was arriving by air. Two batteries plus one platoon of 75-mm. howitzers; two 105-mm., and two 155-mm. howitzers, were landed. All except two pieces with GALAHAD were kept under headquarters control. During the siege they fired 600 tons of ammunition, very rarely with massed fire. Boatner renewed his offensive on June 13th, sending almost all his forces to attack the town from all directions. The American-Chinese forces would struggle to make much progress against General Minakami's defenses. Utilizing a system of night attack and daylight defense, heavy casualties were consequently inflicted on the enemy and large supplies of weapons and ammunition were captured and used in the defense of the city. On June 12th the Japanese hit a platoon of K Company, New GALAHAD, so hard that the company broke and re-formed on the L Company line. The portion of the Japanese thrust that hit the perimeter next to the river made most of the men "take off," but two stayed in place and repelled the Japanese with an automatic rifle and a machine gun. To the west of this little break the Japanese worked their way in close but were stopped by grenades and small arms fire. As a result of the attacks, however, the 3rd Galahad Battalion had cut the Maingna ferry road and reached the Irrawaddy north of Myitkyina by June 17th, with the 150th and 88th Regiments also gaining up to 200 yards. The allies needed to capture the Myitkyina-Mogaung-Sumprabum road junction; but for such few gains, Boatner had lost many men and thus had to stop his attacks on June 18th. Stilwell ordered the end of all infantry attacks. Boatner replied that he would stop attacking Japanese positions until ". . . our troops are steadied and a favorable opportunity presents itself." There was reason for the troops to need steadying. A and B companies, 209th Engineers, were cut off from their main body by infiltrating Japanese. Trying to close in on them, Company C and Headquarters and Service Companies were in turn halted by Japanese. The condition of A and B Companies became critical during June 14th, for they had only one meal with them. Two of their men managed to work their way back to the block on the Sumprabum Road with news of their plight, but enemy small arms fire prevented airdrops. The isolated companies finally made their way back in small groups to the rest of the battalion over 15 and 16 June. The 3rd Battalion of GALAHAD reported trouble in effecting reorganization and enforcing orders. The Americans were not alone in their problems. Two companies of the Chinese 2/42nd which had made a small penetration into the Japanese lines on 14 June were wiped out by counterattack that night. These setbacks emphasized the nature of the Myitkyina fighting. The Allies held a ring of battalion and regimental strongpoints enclosing a similar Japanese system. Though the Allied strongpoints were close enough for the troops in one to sortie to the aid of another should that be needed, they were not so close that interlocking fire could be put down to close the gaps. Consequently, there was plenty of room for maneuver and ambush, and the inexperienced engineers and New GALAHAD troops often suffered at the hands of General Minakami's veterans. On the other hand, the Myitkyina Garrison did not emerge entirely unscathed, as they too suffered heavily, losing approximately 1000 men during the month of June alone. But that is all for today with the CBI theater as we now need to head over to New Guinea to start the Battle of Lone Tree Hill. After the fall of Wakde, General Tagami had sent Colonel Yoshino Naoyasu's 223rd Regiment to cross the Tor River inland to attack Arare while Colonel Matsuyama Soemon's 224th Regiment attacked the Toem area from the other side. In the meantime, General Doe's 163rd Regiment patrolled across Tementoe Creek and the Tor River, encountering heavy Japanese resistance at Maffin but successfully repelling some enemy counterattacks. While the 163rd strengthened its defenses on the Tor and at Arare, Doe would also see the arrival of Colonel Prugh Herndon's 158th Regiment on May 21. General Krueger's plan was to use this regiment in a vigorous overland drive toward Sarmi, aimed at throwing the enemy into the defensive and therefore securing the Wakde area. This decision was based upon scanty and incomplete information concerning Japanese strength and dispositions. The Japanese had no intention of abandoning Sarmi and the two airstrips between the town and the Tor without a desperate struggle.  The Americans were also finalizing their plans for an operation against Biak, codenamed Hurricane. General Fuller's plan was to land the 186th Regiment in the Bosnek area at 7:45 on May 27th to secure the Green Beaches and its two jetties. Once the two jetties were secured, LCI's bearing the 162nd Infantry, supporting troops, and the task force reserve were to move inshore and unload. LST's were also to move to the jetties when the beach area surrounding them had been cleared by the 186th Infantry. LCM's bearing artillery, tanks, and engineering equipment were to move to the beaches as soon as channels through the coral were found or made, or to the jetties in waves following the 186th Infantry's assault companies. As soon as it reorganized ashore, the 162nd Infantry was to advance rapidly west along the coast from Bosnek to seize the three airdromes. This drive was to be supported by eight tanks of the 603rd Tank Company and the 146th Field Artillery Battalion. The fields were to be repaired quickly to accommodate one fighter group and then expanded to receive an additional fighter group, a heavy bomber group, a reconnaissance group, a night fighter squadron, and one photo reconnaissance squadron. Mokmer Drome was to be the first field developed. Brigadier-General Edwin Patrick would also replace Doe in command of the Tornado Task Force, as the latter would resume its duties as assistant commander of the 41st Division. Admiral Fechteler's Task Force 77 was to provide naval support and cover the assault shipping. Naval fire support was to begin at H minus 45 minutes, 6:30. From that time until H Hour, cruisers and destroyers were to expend 400 rounds of 8-inch, 1,000 rounds of 6-inch, 3,740 rounds of 5-inch, and 1,000 rounds of 4.7-inch ammunition on targets in the airfield area west of the landing beaches. After H Hour the cruisers were to continue intermittent fire on the airfields, bombard targets of opportunity, and respond to calls for support from the forces ashore. Because there were many known or suspected Japanese gun emplacements along the south shore of Biak, counterbattery fire was to take precedence over all other types of fire. Bombardment of the landing beaches was also to begin at H minus 45 minutes. Five destroyers were to bombard the beaches and adjacent areas until H minus 30 minutes, when they were to move westward to join the cruisers firing on the airfield area. Then four other destroyers were to continue beach bombardment until H minus 3 minutes. Total ammunition allowance for beach bombardment was 4,900 rounds of 5-inch and 4.7-inch shells, while 40-mm. and 20-mm. ammunition was to be expended at the discretion of individual ship commanders. Rocket and automatic weapons fire from three rocket-equipped LCI's and two SC's was to provide close support for the assault waves. This fire was to begin at H minus 5 minutes and was to last until H Hour or until the initial wave was safely ashore. Meanwhile General Kenney would toss 52 B-24's to bomb the beaches just before the landings. Additionally, medium bombers and fighters from 5th Air Force would cover the force from the air; and from May 17th onwards, the bombings on Biak's airfields would increase sharply in violence to soften up its defenses. As elsewhere along the absolute defense zone perimeter, primary emphasis was laid upon the construction of airfields. Between December 1943 and the enemy invasion of Hollandia in April 1944, two of three projected fields on southern Biak were completed and put into operational use by planes of the Navy's 23rd Air Flotilla. Their usefulness ended almost immediately, however, when the enemy's vastly superior air forces began operating from Hollandia bases. As in the Wakde-Sarmi sector, the concentration of effort on airfield construction until the Hollandia invasion resulted in dangerously delaying the preparation of ground defenses against enemy amphibious attack. In the five weeks which elapsed between the Hollandia and Biak invasions, the Biak garrison forces, under able leadership and by dint of desperate effort, succeeded in organizing a system of strong cave positions, which proved highly effective after the enemy landing. However, time, equipment and manpower were so short that defensive preparations could not entirely be completed. Some 15-cm naval guns, brought to Biak immediately after the Hollandia invasion to strengthen the coast defenses, were still unmounted when the island was attacked. On May 23rd, the 158th advanced west from the Tor River Bridgehead. The advance of Company L met increasingly strong resistance. Japanese defenses were centered around three small, brush-bordered lakes near the beach about 1,800 yards west of the Tor. The rest of the 3rd Battalion, 158th Infantry, across the Tor before 1130, quickly moved forward to assist Company L, which had been pinned down along the main coastal track west of the lakes by Japanese machine gun and rifle fire. Company K pushed up to the left flank of Company L, while Company I moved toward L's rear. With the aid of mortar fire from the 81-mm. weapons of Company M, Companies K and L were able to push gradually forward during the afternoon, advancing on a front about 400 yards wide. Finding that the attack was not progressing as rapidly as he had expected, Colonel Herndon ordered his 1st Battalion across the Tor. The 1st Battalion did not start moving until 1400 and could not get far enough forward to join the attack before dark. Tanks would probably have been of great help to the 3rd Battalion, but by the time the mediums of the 1st Platoon, 603rd Tank Company, moved across the Tor, the forward infantry troops had already halted for the night. In the end, Companies L and K would dig in for the night across the main coastal track about 400 yards east of Maffin.  The following morning, after an ineffective mortar and artillery bombardment, Herndon resumed the attack. Despite the lack of extended artillery support, Companies K and L moved out as planned at 7:30. Company L, on the right, advanced along the beach encountering only scattered rifle fire but Company K, on the main road, had hardly started when Japanese machine gun and rifle fire from concealed positions in a wooded area on the left front halted its advance. Unable to gain any ground, Company K called for tank support. Two tanks, together with a flamethrower detachment from Company B of the 27th Engineers, arrived at Company K's lines about 1000. With the flamethrowers and tanks blasting the way, the infantrymen overran the Japanese defenses, killing ten of the enemy and capturing two machine guns. The remainder of the Japanese force, probably originally some forty men strong, disappeared into the jungle south of the road, whence scattered rifle fire continued to harass Company K. Company L reached the outskirts of Maffin No. 1 about 1400. The movement had been slow, not as a result of Japanese opposition but because the battalion commander did not believe it prudent for Company L to advance far beyond Company K. Despite the return of two companies, most of Colonel Kato's engineers would have to withdraw behind the Tirfoam River against such heavy firepower. Captain Saito's reconnaissance unit, meanwhile, retreated to the jungles south of Maffin alongside one engineer company, which was under Kato himself.  Over the Tirfoam River, however, the Americans were again stopped by the tenacious engineers, which again forced Herndon to request tank support. As the tanks moved into position elements of the Right Sector Force, comprising Captain Saito's men of the 1st Battalion, 224th Infantry and an engineer company, charged out of the jungle. The Japanese were under Colonel Kato, Right Sector Force commander, who was killed as he personally led a small detachment against the American tanks. The enemy was quickly thrown back with heavy losses by the combined fire of the four tanks and Company L's riflemen and machine gunners. However, under cover of their infantry attack, the Japanese had dragged a 37-mm. anti-tank gun forward out of the jungle. As the enemy infantrymen withdrew to the southwest after the death of Colonel Kato, the anti-tank gun opened fire. It was soon destroyed and its crew killed, but not before three of the American tanks had been so damaged that they had to be withdrawn for repairs. Facing such heavy resistance, Herndon then sent his 1st Battalion to carry out a deep envelopment to the south across the Tirfoam, yet these troops would similarly be unable to break through by nightfall. After killing 28 men and wounded 75 others, Kato's force allowed Tagami to dispatch the 2nd Battalion, 223rd Regiment to reinforce the Ilier Mountains line. On the morning of May 25, Major Matsuoka Yasake also arrived there with an infantry company to assume command of the remainder of Kato's forces. Meanwnhile Yoshino had crossed the Tor River, 3 days behind schedule and to the east, Matsyuama was assembling on the right bank of the Tementoe River. Herndon relieved his 1st Battalion with the 3rd and then pushed west with the 2nd Battalion following behind them. The next objective for the 158th was Long Tree Hill. Lone Tree Hill, known to the Japanese as Mt. Ilier,  had been named for a single tree which was depicted on its crest by the map then employed by the Americans. Actually, the hill's coral mass was covered with dense rainforest and jungle undergrowth. Lone Tree Hill was about 175 feet high, 1200 yards long north to south, and 1100 yards wide east to west. The north side dropped steeply to a rocky shore on Maffin Bay. The hill's eastern slope was fronted by a short, violently twisting stream which was promptly dubbed the "Snaky River" by the 158th Regiment. The main road curved away from the beach to pass south of the Snaky River and Lone Tree Hill through a narrow defile. The southern side of this defile was formed by two noses of Mt. Saksin , a terrain feature about 100 feet higher than Lone Tree Hill. The more westerly of these noses was named "Hill 225", known to the Japanese as Mt. Sento after its height in feet. No name was given to the eastern ridge line, which pointed toward Lone Tree Hill from the southeast. There was a small native village at the eastern entrance to the defile and another at the pass's western outlet. Mt. Saksin was a name given to an indefinitely outlined hill mass which forms the northern extremity of the Irier Mountains, extending inland from the coast at Lone Tree Hill. The name Saksin was specifically applied to a prominent peak about 2,000 yards due south of Lone Tree. On or about May 23rd General Tagami had moved his headquarters into the Mt. Saksin area, apparently on the southwest side of the central peak.  Herndon forces continued the advance through the abandoned enemy positions, albeit without tank support, for they could not cross the fragile Tirfoam bridge. In the afternoon, the Americans were finally halted below the southernmost bend of the Snaky River, subjected to heavy machine-gun fire and an intermittent artillery bombardment. General Patrick, who had succeeded to the command of the TORNADO Task Force during the morning, was informed of the opposition encountered by the 1st Battalion. He ordered the advance stopped for the night and instructed the 158th Infantry to remain well east of the Snaky River so that American artillery could register on the native village and the defile without endangering the forward troops. Harassed by a few artillery shells, which by now had been recognized as originating from Japanese 70-mm. or 75-mm. weapons, the 1st Battalion pulled back about 500 yards east of the Snaky. A perimeter was set up with the battalion's left resting on the road and its right on the beach. The 2nd Battalion established a series of company perimeters back along the road to the east. Casualties for the day had been 22 men killed and 26 wounded, almost all in the 1st Battalion, while about 50 Japanese had been killed. When the attack orders for the day had been issued, it had been hoped that the 1st Battalion could reach the top of Lone Tree Hill before nightfall. Since the unexpectedly strong enemy opposition had prevented the realization of this hope, plans were made to continue the advance westward on the 26th. The ultimate objective was the east bank of the Woske River, 2,000 yards west of Lone Tree Hill, and the intermediate objective was the native village at the eastern entrance to the defile. The advance was to be preceded by naval shelling of the northern slopes of Lone Tree Hill from 6:30 to 7:00. A fifteen-minute artillery preparation was also to precede the advance, and the infantry was to start moving at 8:45. The next morning, naval fire started ten minutes late. Two destroyers lying offshore shelled the northern slopes of Lone Tree Hill and the Maffin Bay area, firing on known or suspected enemy defensive positions and assembly points. After a twenty-minute bombardment the two support vessels withdrew. Artillery fire did not begin until 8:30. The time lag gave the Japanese ample opportunity to prepare for the infantry attack which had been heralded by the destroyer fire. The artillery, aiming its shells into the defile and against the eastern slopes of Lone Tree Hill, ceased firing about 8:45. A few moments later the 1st Battalion, 158th Infantry, Company B again leading, started moving westward. The infantry's line of departure was nearly 1,000 yards east of the village at the southeast foot of Lone Tree Hill, and the advance had to be slow because the road ran through heavily jungled terrain. The enemy therefore had sufficient time to reoccupy positions in the defile and on Lone Tree Hill which might have been vacated during the American artillery barrage. The value of both the naval and artillery bombardment had been lost. Herndon's 1st Battalion moved once again against Lone Tree Hill. Company B moved forward to the point at which it had been held up the previous afternoon and was again stopped--this time by fire from the southeastern corner of Lone Tree Hill. Company D's heavy machine guns were brought up to spray a densely wooded area in front of the point rifle platoon. The fire dispersed the Japanese riflemen, and Company B moved forward again. Less than 100 yards of ground had been gained when the company again encountered machine gun and mortar fire originating in the native village. Company A, initially off the road to the right rear of Company B, turned north to the mouth of the Snaky River. One platoon crossed at the river mouth at 1030 but was quickly forced back to the east bank by Japanese machine gun fire from the rocky beach below the north face of Lone Tree Hill. Artillery support was called for, supplied, and proved successful in stopping the enemy fire, and about 1:50 all Company A crossed the Snaky. Orders were to move down the west side of that stream to establish contact with Company B and to send one platoon up the eastern slope of Lone Tree Hill to probe enemy positions. Other efforts were meanwhile being made to scatter the Japanese opposing Company B. Company E moved up to the left flank of Company B and on the south side of the main road. The combined efforts of the two rifle companies proved insufficient to dislodge the Japanese from their positions at the eastern entrance to the defile, and the enemy fire forced the American units to seek cover. Company F was therefore ordered to pass through B's left flank and proceed to Hill 225 to take the Japanese positions from the rear. Company F's attack could not be started before dark and Company A, moving up the west side of the Snaky, was unable to relieve much of the pressure on Company B. Finally, Company A was forced for a second time to withdraw to the east bank of the river as a result of enemy fire from Lone Tree Hill. Tanks would have been of great help to Company B, but the bridge over the Tirfoam could not bear their weight, and the road west of the stream was in such disrepair that tanks probably could not have negotiated it. Only Company A would manage to cross the Snaky River through much effort, yet it would be unable to relieve the pressure from Company B's front and would have to retreat by the end of the day. To prevent further casualties from being inflicted by Japanese patrols, which were expected to roam around the flanks of the forward elements during the night, a semicircular perimeter was ultimately established. Although Herndon's attacks on May 26th had been completely unsuccessful, the 158th Regiment had located and probed some of the principal Japanese defenses in the area and could now be ready to launch a more effective assault.  I would like to take this time to remind you all that this podcast is only made possible through the efforts of Kings and Generals over at Youtube. Please go subscribe to Kings and Generals over at Youtube and to continue helping us produce this content please check out www.patreon.com/kingsandgenerals. If you are still hungry after that, give my personal channel a look over at The Pacific War Channel at Youtube, it would mean a lot to me. General Vinegar Joe was experiencing some major setbacks, but it looked like Myitkyina was as good as his. However the American officers' reports of how difficult the going was for the Chinese was quite disheartening. Over in New Guinea, the allies were yet again faced with a great obstacle, now in the form of One Tree Hill.