Comparative spychology

Frank E. Lutz, « Caddys-Fly Larvae  as Mason and Builders, Flexibility in the Case-making Behavior of Caddis-Fly Larvae », New York,  Natural History, The Journal of the American Museum of Natural History,  vol. 30, 1930, pp. 275- 281.

Caddis  flies, although classified  as a distinct order, Trichoptera, seem to be very close relatives of primitive moths. Their larvae are, for the most part, aquatic, and typically make cases within which they live. The simplest of these cases are merely thin silken envelopes in which silk or other foreign material may be enmeshed. However, perhaps the majority of species, certainly the majority of species whose life history has been recorded, build cases by cementing together bits of foreign material with silk.

The interesting thing about these cases is that each species has a characteristic choice of material and a definite style of architecture. The choice of material is, naturally, limited to the things found in the particular aquatic environment which the species inhabits: vegetable matter, either living or dead, and harder, unchewable substances such as sand, small pebbles, and snail-shells. The architecture, taking the group as a whole, is most diverse, ranging from mere agglomerations of almost anything to a spiral case made of fine grains of sand so neatly and regularly fitted together that it was originally described as the shell of a snail new to science.

But, whatever the style of architecture or the choice of material, it is usually characteristic of a given species, sometimes even of a genus or family.

There is a correlation between the kind of case which a species makes and the swiftness of the water in which it lives. Thus, the species which make a « log cabin » case of small sticks placed crosswise, as shown on this page, lives in relatively still water such as ponds, while one that fashions a « masonry mosaic » of pebbles (page 278) lives in streams. This is usually regarded as a wonderful adaptation of behavior to environment, but perhaps, since we do not know whether it was environment or behavior which changed, it would be better to say that it is an adaptation between behavior and environment. Certainly a larva that made a log-cabin case in a stream would soon be washed down to still water. It s not quite so clear why species that make masonry cases should not live in still water and, in fact, they sometimes do, but possibly one reason for their not being abundant in such places is that pebbles are not so apt to be available.  Since the caddis flies are a relatively old group, being well represented at least as early as the Oligocene, and since  certain types of cases are also characteristic of genera or even of families based entirely on anatomical characters of adult insects, it is quite probable that a species which builds a given kind of case now has built nothing but that kind of case for thousands or even millions of years. What would a caddis larva do if it could not get material like that to which it has been accustomed and, further- more, could it live without a case as do other aquatic insect larvae?

The answer to the first of these questions has been partly known for some time. Larvae dispossessed of their cases have been given bits of mica from which to make new cases in order that students might be able to watch the actions of the larvae within the cases. Ostwald, in particular, got larvae to build with a great variety of unusual substances. In trying to find the answer to the second of the questions, I encountered a difficulty which had probably brought failure to others who said nothing about it, but I did find a few facts which seem to have an interesting bearing on the first question. These larvae are vegetarians and, if given vegetable matter for food, nearly all, even the masons, not only could but did



A larva from a « log-cabin » case in one which it started by using pieces cut from Utricularia and finished, when it could get them, with sea-urchin spines use it for making cases. Possibl}’ by starving them sufficiently and then giving them only exceedingly small pieces of food at a time, one could force them to go without shelter rather than go hungry, or one could concentrate on the less adaptable; but, everything considered, the first question seemed to offer more interesting possibilities.

For example, four log-cabin builders were dispossessed of their cases and put into an aquarium with nothing but water and sea-sand with a little alga for food. This experiment was being carried on at the Mount Desert Island Biological Laboratory and the available sea-sand was made up largely of fine pieces of broken shells. By the next day one larva had made a nice case of shells and sea- urchin spines; one used some of these, but mostly alga; and two were still naked. The following day three had shell and

spine cases, the alga that one had used having apparently been worn or eaten off,

and one still had no case, but there were two partly made and then abandoned

cases in the aquarium. The material normally used by the larvae which build the log-cabin type of case consists of more or less cylindrical pieces of soft, partly decayed twigs or, in this Maine locality, of spruce needles  which had fallen into the water.

These are placed very definitely in a transverse position, with reference to the larva’s body. When I put the sea-sand into the aquarium, I did not notice that it contained a small amount of sea-urchin spines. It was, therefore, a matter of considerable interest that the larvae sorted over the sand and picked up here and there for use such a large proportion of the relatively rare objects which came nearest to being the shape of their normal material. That this was not because they could not use the flat pieces of broken shell is shown by the fact that they did use a few of such pieces and, in the case of one larva, it started with the exceedingly thin and flexible strands of alga, although it later changed to sea-urchin spines and some broken shell.

This seeming exercise of choice, a possible evidence of real mental preference, was further illustrated in another experiment in which dispossessed log- cabin builders were first put into an aquarium with nothing but water and Utricularia, a delicate plant bearing curious  »bladders » or pockets which entrap microscopic organisms. Nearly all of the larvae rather quickly made cases out of pieces of the plant which they bit off.

In order to understand more fully subsequent developments it should be remembered that such a larva constructs its case by making a narrow ring through which it puts its head and then it widens the ring by adding material to the front side until the ring has become a cylinder as long as, or longer than, the larva’s body. New material may occasionally be added to the side of a case but, when the case is too short or for some other reason is not satisfactory, matters are usually adjusted by adding to the front end.

Possibly the fact that the cases made by log-cabin builders, when they were forced to use shells and spines, were longer than normal was due to the insects’ disatisfaction with the new cases and their attempt to make them better by continued construction.

Be that as it may, the log-cabin builders in their Utricularia cases did not live

peacefully together. One would come up behind another and steal a bit of plant

from the case of the second in order either to eat it or to add that already-cut

piece to its own case instead of cutting another piece for itself. After allowing

this shifting of material from the rear end of one case to the front end of another to

go on for a day or two, I put sea sand into the aquarium. Each larva had at that

time a reasonably good, although rather short, Utricularia case. However, each

began picking up sea-urchin spines and adding them to tho front of its case, occasionally, as though b}- mistake, fastening one to the side. The figure at the  top of page 275 shows such an instance. Eventually, because the Utricularia was either worn or eaten off, most larvae had complete cases of sea-urchin spines. Possibly the scarcity- of broken shell in these cases, as contrasted with those made by naked larvae which started with nothing but sea- sand and a little alga, is to be explained by the fact that these larvae already had fairly satisfactory cases and, so, made a more leisurely and careful selection of material. I quite realize the danger of such a suggestion with its implication that larvae of lowly creatures think. Possibly the preference shown for spines over broken shell may have a purely mechanical explanation in the shape of the tarsal claws with which the building material is handled, but I am at present unable to suggest a purely mechanical explanation for this preference being more completely manifested when the larvae already had cases.

Larvae that normally make a somewhat curved case of fine grains of sand (see

figure top of page 276, in which a small bit is broken off from the hind end of the case)

were dispossessed and given only the coarser bits of broken shell that had been sifted from sea-sand. In view of what has already been said, it is not surprising that they made cases of this new material. The experiment is mentioned here to show by the figure at the bottom of page 276 how well they handled the larger blocks, doing almost, if not quite, as well as a species which normally (page 278) uses. This case was made from « log-cabin » material by a larva which normally uses only sand  relatively large pebbles, the sort of case that was referred to above as « masonry mosaic. »

A more surprising result came when larvae, dispossessed of their fine-sand cases, were given only spruce leaves and small sticks taken from log-cabin cases.

Some with very little hesitation, others with more, made cases such as the one shown on this page. They did not arrange the material transversely, as do the log-cabin builders, but more or less longitudinally, as do many other species of caddis larvae. There is no warrant for making much of the fact that this is a better arrangement of material forspecies which, like these sand-builders, live in relatively swift water, but we can say very definitely that what was probably the first attempt this species had ever made in constructing cases from such material resulted in a neat and apparently very satisfactory shelter of a type quite different from its previous experience but entirely normal for other species.


Masons such as the sand-builders just discussed or those that use small pebbles

never cut material for their cases; they merely pick up pieces of suitable size.

Therefore, the following experiment seems worth noting. On August 2, 1 dispossessed

three sand-builders and gave them both living Utricularia and material taken from

log-cabin cases. The next day one had made a case of spruce leaves, one used

these and some fine debris, and one used pruce leaves but also pieces of Utricularia which it had cut off from the living plant. By August 6 this third larva had

added sticks on the front of its case and had lost most of the Utricularia from the rear, probably by theft. Five days later, when all three had stick cases, I dispossessed two of them and put them in a new aquarium with only Utricularia in the water. In the same aquarium Iput two fresh sand-builders : one which I had robbed of about half of its case, leaving it in the much-too-short other half, and one which I entirely dispossessed. I do not know how things fared by the next day, but on August 13 the three which were entirely dispossessed had good cases made of pieces of Utricularia which they had cut from the plant; the one with a too-short case had done nothing about it. Thus matters stood for several days, but later one larva disappeared — I am not sure which one or how. There was considerable fighting and stealing of material from each other’s case until, when the experiment was stopped on August 26, one larva was naked (possibly the one which had had a piece of a sand case) and the cases of the other two consisted

chiefly of the silken lining with small bits of Utricularia and debris still attached.

Apparently cases made of material which can be eaten — gingerbread houses,

as it were — are not extremely satisfactory, except to the eater, but the

experiment suggests the possibility that these larvae, although not endowed by

training and probably not by instinct with the notion of using vegetable material

for their cases, can not only use such material but can do the, to them, unprecedented thing of cutting the case-building material into usable lengths. So far as this experiment goes the last statement is suggested as mere  » a possibility, there being the other possibility that they cut off pieces from the Utricularia for food and then, having the pieces, used them for making a case. However, the next and, so far as these notes are concerned, last experiment to be presented, seems to make the suggestion more probable.


Larvae that make tile-like cases of relatively large pebbles, the « masonry mosaic » type shown on this page, were dispossessed and given the usual variety » of unusual material. They managed them all so well that I finally gave some freshly dispossessed ones large decayed leaves. Certainly these soft flat objects, too large to be fastened to a case without cutting, were very different from small, hard, round stones which are not to be cut. A creature whose ancestors to remote generations had never used anything but pebbles might not be expected to « think of » (« respond by » in more technical language) using decayed leaves or to be able to use them whether it really thought about the matter or not. All that 1 can say is that it did cut pieces out of the leaves and from them case shown on page 278.


One thing that should be mentioned again has been touched upon from time to time in these notes. Not all of the individuals of any of the species, which were given these tests solved the problems. Quite probably most of those in any given experiment were brothers and sisters, or at least closely related, as they were collected together, but some would be immediately successful with the new material, others were slower or less successful; some tried for hours and failed, leaving partly finished cases behind them, and some did not try. That the last fact was not due to lack of either silk or case-building urge was shown by their quickly making normal cases when given normal material. It almost looks like a variability of intelligence or ingenuity or something of the sort that many students suppose insects lack.

On the whole, without attempting to define either instinct or intelligence but selecting « behavior » as being the safer term, it seems clear that the case-building behavior of caddis-fly larvae is at least a rather flexible affair. Other more debatable points will occur to those familiar with comparative psychology.