Following up my rainbow thread, I starting thinking about the tertiary bow and indeed you can see it on a semi-transparent screen placed behind the glass sphere (i.e. in the same direction as the incident rays). The vertical metal post in the photo blocks the focussed light.
So the photo above shows simultaneously primary, secondary, and tertiary bows! Very nice. For completeness, I am going to now show some raytracing pictures for this case of all three caustics. First with one ray comparing ray paths in water and glass.

Preparing my 2025 Day of Light outreach lectures, I bought a glass sphere and made a rainbow. But I saw some odd things, and so here’s a thread with some photos & raytracing pictures to try to explain what's going on. First the basic setup showing the primary bow. Where is the secondary bow? It took longer than it should have to find it, and it turns out it’s a long way away, nearly perpendicular. An angled screen helps and the secondary is then clearly seen with inverted colours as expected. But why is it so far away from the primary?

Feynman's QED: The Strange Theory of Light and Matter, was first published in 1985. In response to two completely unconnected queries in the last few weeks, here's a thread on how the lectures on which the book were based were first given in New Zealand. The story begins in 1979 when Feynman gave the Sir Douglas Robb Lectures at the University of Auckland in New Zealand. These talks were a testing ground for his 1983 UCLA Alix G. Mautner Memorial lectures, widely considered to be the basis of the 1985 book.

This year is the 600th birthday of the University of Franche-Comté, the 10th university created in France in 1423. For @IDLofficial I gave a talk on optics history here since the science faculty was created in 1845. First batch of tweets follow; text in English, slides in French. The story begins with our project with @SSAC_Univfc to save the lab archives that date back over 100 years. We found many old cans of photographic negatives from the 1970s and one was especially intriguing – who were these "ancestors"?

Something different! Delighted to announce that a paper I wrote 22 years ago (!) on a supposed portrait of English scientist Thomas Harriot (c1560-1621) is finally online: TL;DR: Sorry but there's no real evidence that the portrait is Harriot ... 🧵 hal.archives-ouvertes.fr/hal-03839673
Harriot was a polymath. He produced the first telescopic drawing of the moon before Galileo, he discovered the law of refraction before Snell or Descartes, and he explored Virginia, learning Algonquian to translate. It would be nice to know what he looked like.

To kick off the week, here is an updated (and long!) thread on the history of nonlinear optics. First a real surprise! The explicit use of the terminology “nonlinear optics” can be traced back to Erwin Schrödinger in 1942. Yes you read that right. Schrödinger himself! Although this paper isn’t really what we would describe today as non-linear optics. Rather it describes “vacuum light-light scattering” or nonlinear QED. But the wording Schrödinger used definitely sounds familiar! It builds on earlier work by Born, Infeld, Euler.

I was honored to speak at Moscow State University last week for the 60th anniversary of laser nonlinear optics. But in fact, the first nonlinear effects in optics were observed in the pre-laser era by Vavilov in Moscow in 1926! Here are some slides from my talk (thread). Sergey Ivanovich Vavilov (Серге́й Ива́нович Вави́лов) was a giant of physics. He invented the term "nonlinear optics" in his 1950 book & was co-discoverer of Cerenkov radiation. Regrettably he died before the Nobel Prize for Cerenkov radiation was awarded in 1958.

An important anniversary next week! 20 years since I left @UoA_Physics in beautiful Aotearoa to live in beautiful Besançon. In the best academic tradition, must be time for a 20 year Activity Report! Thread follows: @fc_univ @FemtoSt @INSIS_CNRS @CNRS_Centre_Est Important caveat. Don’t believe for a second that everything ran smoothly! Many failures - rejected papers & funding, most ideas went nowhere, many mistakes. But you keep at it and with LOTS of help you somehow get somewhere in the end.

As promised a thread (17 tweets) with a selection of photos from the Live Science lecture last night. 200 students, 25+ demos & a team of 7 to setup. Thanks to @CocoLapre for the photos & a full list of thanks to everyone is at the end. First up dispersion & rainbows. After breaking up white light, we put it back together. Great chance to talk about how flat screen displays work at this point.

Lecture demonstrations for a Saturday morning. First - a rainbow. Very simple with a projector, card and a flask. Methyl Cinnimate is better than water for doing this as it has a higher refractive index. And complementing a rainbow is the classic dispersion of white light in a prism demo - colors seem to have saturated on the card but not on the reflection on the table.