2l6 



SCIENCE. 



[Vol. XXn. No. 559 



over which water flows with considerable speed are liter- 

 ally blackened with it. Since the note referred to was 

 published I have observed that these latter will when dis- 

 turbed let themselves loose in the current and then shoot 

 down stream emitting their threads at the same time so 

 that they can check their descent and secure a fresh hold 

 on the rocks, perhaps to return along the thread to their 

 first position. By closing the blades of my forceps over 

 the rocks I have repeatedly drawn out a string of these 

 larvae, each one suspended by the thread it had let out as 

 it floated down stream. 



In addition to this thread-spinning habit the cave larvae 

 have peculiarities of structure which render them worthy 

 of careful study. I have already described two of them, 

 and have collected several others in Kentucky. All are 

 more or less vermiform, being long, slender, cylindrical, 

 generally translucent, so that the internal organs show 

 more or less distinctly through the body wall. The re- 

 semblance to a small Lumbricoid worm is heightened by 

 the fact that their bodies are coated with a slime, the de- 

 rivation of which is uncertain, but which is probably not 



Fig. 2. 

 derived from the glands engaged in secreting the thread. 

 There are common features also in the structure of the 

 head and mouth-parts, and in the presence of a singular 

 convex area above the base of the mandible resembling a 

 very large ocellus. In several of those examined this is 

 of enormous size, and gives the head a most bizarre ap- 

 pearance. 



That they are Dipterous larvae is sufficiently evident 

 from their resemblance in general structure to larval 

 Seiara. The recent discovery of the pupa of one of the 

 species, with wing pads and halteres clearly apparent, con- 

 firms the opinion I had reached in this regard. AVhile 

 engaged in attempting to rear the adult of this species I 

 received additional proof in the shape of a letter, quoted 

 from below, from that most excellent observer and collec- 

 tor, Mr. H. G. Hubbard, together with three stages of a 

 closely related species which he discovered some years 

 ago in a cave in Jamaica. The larva of this species is 

 closely allied to one found by me during the past August 

 living in hammock-like webs slung across depressions on 

 the under side of stones and lumps of earth. The latter 

 species was taken in a small cave near Lexington, and has 

 afforded me an opportunity to observe more closely the 

 product of the spinning glands of these interesting in- 

 sects, and to watch the larva while making its web. This 

 larva shows the same attachment for its web as does the 

 species previously described. In one instance an example 

 was compelled by particularly rough treatment to creep 

 to the earth at one side of its web, where it remained 

 drawn up in an uncomfortable position, but turned 

 promptly when left unmolested for a moment and made 

 its way back on the web again. Three living examples 

 were at one time thrown into a watch glass of water pre- 

 paratory to killing them in an extended condition, when 

 every one fastened itself to the bottom by pouring out 

 the glutinous material from its mouth and then began to 

 wriggle like an uncomfortable earthworm, always with 

 the whole length of the body free from the glass. In 

 this case the slime coating of the body showed no ten- 



dency to glue the body down, whereas the matter from 

 the glands opening at the mouth retained all its adhesive 

 proporties — an evidence that the slime is of different 

 origin, and is produced for a different j)urpose. 



Since Sept. 3 a larva of this sort has been kept alive 

 in a bottle. In the bottom of this is about half an inch of 

 earth. The larva si^ends most of the time in the empty 

 upper part and makes its way about in this space, build- 

 ing a web as it goes, with surprising rapidity. It is often 

 fully two inches from the earth and very rarely touches 

 the side of the bottle with its body. When engaged in 

 web-building, it sways the forward part of the body from 

 side to side until it strikes some object, when the thread 

 is attached by a touch, and as the head draws away is 

 seen to be connected with that underlying the whole length 

 of the body. When first drawn out these threads apjsear 

 under a hand lens as smooth and dry as any spider's web. 

 The central strand upon which the larva usually lies, 

 however, has a good deal of slime along it, forming tri- 

 angular masses at the points of divergence of lateral 

 threads. When they have been used for some time the 

 lateral threads of a web may also show slime upon them 

 in the form of minute scattered spherical droplets (See 

 figure). As far as I can determine all this slime comes 

 from the surface of the body. Occasionally a portion of 

 the body has been seen to come in contact with the bot- 

 tle, where a slimy trail nearly as wide aa the body was 

 left on the glass. If this slime had the properties of the 

 glutinous material of which the thread is made the larva 

 would have difficulty in getting about. On the contrary 

 it is rather fluid, and the droplets left along the strand 

 can be seen to be drawn up by the force of capillarity as 

 the tip of the body passes them. 



Fig. 3. 



These larvae live concealed in damp situations and it 

 may be, as suggested by Mi-. Hubbard, that the threads 

 do not become perfectly dry. They are so fine and deli- 

 cate that it would be difficult to determine this matter. 

 The thread-making larva previously discribed in Science 

 is at all times completely exposed on the rocks. I have 

 had no recent opportunity to examine its threads, but the 

 impression I have of those seen last sjjring is that they 

 were dry. But the question w hether or not the threads 

 of these larvae are completely dry has nothing to do with 

 that concerning their essential nature. If silk must be 

 chemically dry, then of course the thread of larval Simu- 

 lium is not silk. It is not the product of a gland having 

 to do with digestion. It is not a trail of slime left from 

 the surface of the body. It is a special product, used by 

 these larvae exactly as the silkworm uses the product of its 

 sericteria (even to enclosing the pupa in some cases in 

 a very slight approach to cocoon). 



Mr. Hubbard's larvae are very much like the species 

 upon which my observations have been made, and their 

 threads of slime very jarobably have a supporting axis of 

 other material The following quotations are from his 

 letter accompanying the specimens so kindly sent me. I 

 hope to publish descriptions of all the cave species at 



