What might earlier versions of caddisfly shelter-building behavior have looked like?

Question

The caddisfly has an amazing ability to build armor for itself by using a self-produced underwater glue to hold together pebbles: https://www.youtube.com/watch?v=Z3BHrzDHoYo.

How could this behavior and the ability to produce the glue have evolved? The survival advantage of the current level of the caddisfly's ability is clear, but how were the incremental stages advantageous? How could it have developed its armor-building ability without a strong enough glue? And how could it have developed a stronger glue without the armor-building ability that would make it useful?

My guess would be that before it was building a full shell like it does now, it started just by holding a pebble or two to itself. Then perhaps the bodily fluid that is now used as glue gradually increased in stickiness and helped the organism keep hold of the rocks or keep them together. It seems, however, like any modest amount of increase in stickiness that could occur due to mutations wouldn't have been enough to keep the rocks together.

Answer 1

Your intuition seems fairly correct. Mouro et al 2016 show some findings relevant to your question...

1. Caddisflies have been around for a very long time. Much of the fossil record of their existence is of preserved shelter cases rather than the insects themselves.
2. Early cases seem to be made up of more variety of materials, whereas selective cases made of very specific materials are observed more recently in the fossil record and today. This suggests that they first just collected random pebbles and organic materials, and later evolved more refined construction techniques.

As far as evolution of their silk, I think it's important to recognize that these insects are closely related to Lepidoptera, the moths and butterflies. Many Lepidoptera species, like caddisflies, spin silk cocoons as part of their development cycle, so it can be inferred that the silk and many of its properties precedes any construction behavior using other materials.

---

Mouro, L. D., Zatoń, M., Fernandes, A. C., & Waichel, B. L. (2016). Larval cases of caddisfly (Insecta: Trichoptera) affinity in Early Permian marine environments of Gondwana. Scientific reports, 6, 19215.

Answer 2

Beyond what is shown in the fossil record and in genomic analysis, the evolutionary advantage of a phenotypic feature can only be guessed at and its plausibility gauged.

That said, here's a guess that seems plausible: the sticky silk might have first served as a way to tether the caddisfly larvae to stationary rocks in a river current. The tether would have naturally and coincidentally picked up small rock grains, sticks, and the like. If the tether broke, the grains would stay attached to the remaining silk strand and, if "reeled in" by the larva, might have served as makeshift armor as the larva tumbled downstream in the current. This would open up a new adaptive opportunity, channeling adaptations in the direction of improving the shell-constructing behaviors and physiological tweaks that support shell construction.