ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 555 



have at first the forms which they have later on, whereas this is by no 

 means the case ; at first there are nothing but rhomb-shaped spaces, 

 the size of which is gradually increased. 



Closed Poison - glands of Caterpillars.* — Dr. Dimmock states 

 that if a Cecropia caterpillar " be examined carefully, the black 

 spines upon its red, blue, and yellow knobs, or tubercles, will be seen 

 to break easily from the tubercles, and a clear yellow fluid of dis- 

 agi'eeable odour to ooze from each opening left by the injury. By 

 crushing the tubercles with a pair of forceps the same strong odour 

 is very noticeable, and by this mode of treatment one has no difficulty 

 in proving that each tubercle, small or large, blue, yellow, or red, 

 contains the odorous fluid. The red tubercles are seen, in sections 

 cut with the microtome, to be divided into compartments, the cavities 

 of each spine opening into a compartment at its basal end. The 

 spines themselves are quite rigid and very brittle, so that they break 

 away at a slight touch and leave a hole in the tubercle, out of which 

 the odorous fluid pours, pushed by internal pressure. This fluid, 

 which I have not examined carefully, but which I hope later to study 

 chemically, is strongly acid to litmus paper, but causes a purple pre- 

 cipitate in carmine solution." The odour given out by these glands 

 suggests at once their protective function. Similar glands, i, e. with 

 no outlet until one is produced by external agency, are not rare in 

 Bombycid larvaa. Karsten, in 1848, described the anatomy of the 

 poison-glands at the base of the hairs of an American Saturnia. The 

 secretion is "perhaps formic acid or a formate in solution." 



Gills of Insect Larvae.t— G. Macloskie states that it is usual to 

 describe the laminae of the pneumatic gills as containing systems of 

 fine tracheal loops, somewhat after the pattern of a plurality of 

 carbon-wicks in an Edison lamp. In a specimen, however, of the 

 rectal branchiae of the larval LibeUida, which he rolled under the 

 cover-glass, he found that the multitude of tracheal ramifications 

 ended caecally ; all were of about the same length, their extremities 

 recurved within the containing sac, and their tips not all swollen, 

 but rounded off. " As they are elastic, and the closing sac disten- 

 sible, we J think it highly probable that with each water-inspiration 

 the sacs enlarge and the tracheal spray (having air forced in by the 

 forward compression of the large tracheae) spreads out so as to bring 

 the full tide of air close to the tide of water. Leon Dufour seems to 

 have had some process like this in view, when he said that each 

 lamella of the branchia of Potamophilua ' is probably swollen during 

 life by air transmitted by endosmosis.' As we understand the case, 

 the air is injected into the branchiae from the rest of the body by 

 rhythmical contractions, and its gases then communicate endosmoti- 

 cally with those in the tidal waters, so as to secure renovation." The 

 action of the tracheae, Macloskie believes to be tidal rather than due to 

 peripheral capillary circulation ; there being a flux and reflux, rather 

 than a mere circulation of the air. 



* PHyche, 1882 ("4). Airifr. Natural., xviii. (1884) p. 535. 



t Psyche, iv. (1883) pp. 110-2. J Aruer. Natural., xviii. (1884) pp. 534-5. 



2 I' 2 



