NATURAL SCIENCES OP PHILADELPHIA. 29 



species operate to produce functional changes within its economy, 

 the rare occurrence of a cocoon-making species laying aside this 

 hitherto supposed indispensable habit, becomes divested in a 

 measure of the mystery which envelops it, without losing any of 

 its interest. 



To assign a cause for this anomalous natural occurrence shall 

 be my aim. In a series of experiments which I recently conduct- 

 ed of starving larvae to ascertain the effects produced upon their 

 economy, I have been led to observe that in cases effectually ac- 

 complished, there is, besides a remarkable diminution in natural 

 size, in cocoon-producing species, as Cecropia, Polyphemus, etc., 

 a considerable saving of silk. Last year I produced cocoons of 

 Polyphemus, three-quarters of an inch in length, by less than one- 

 half in thickness. The flies which emerged from these cocoons 

 measured If inches in expanse of wings. In the generality of 

 cases males were the result of starvation. 



This last year I fed a number of Cecropia upon the leaves of our 

 ordinary red currant. One of the cocoons which I gathered was 

 less than one inch in length with a thickness of one-half inch. 

 Prompted b}' curiosity, I tore the cocoon open and discovered that 

 the larva had been ichneumoned. In the place of a chiysalis was 

 found the black cylindrical cocoon such as we ordinarily find in such 

 cases. The larva had undergone its skin shedding, as evidenced by 

 the dried inverted skin which was shoved into one corner of the co- 

 coon. In this case, the larva, being weakened and diseased by the 

 ravages of an internal foe, had not the material, or the power, to 

 construct the characteristic type of cocoon. In the case of Polyphe- 

 mus adverted to, scarcity of food, and of an innutritious character 

 in the bargain, had reduced the insect to a condition which ren- 

 dered it powerless to produce the typical form. 



May not the stinting process which necessity compelled -me to 

 adopt with some larvae of Acronycta, have had the effect of deter- 

 mining the acquired habit which has been noticed ? The silk-pro- 

 ducing glands of Acronycta, in consequence of a diminished supply 

 of aliment, were not provided with the requisite material for the 

 manufacture of silk, there being a bare sufficiency of food to 

 prevent starvation to the insect. These silk organs are most de- 

 veloped at the period when the insects approach their pupa state ; 

 but with the larvae of Psychidee, Tortricidse, and Lasiocampidse, 

 they are already active during the earl}'' epochs of life. As the 



