• THE FALL WEB-WORM. 55 



early in October, and busily engaged in inserting their ovipositors 

 through, the tough cocoon into their victim within, i as if the 



cocoons formed early in the season were on an average a little smaller 

 than those formed later. 



"The cocoons of this Apantelcs are of a uniform white color, bn; 

 ceptionally a distinctly yellowish cocoon is found. From these yellow 

 cocoons nothing has so far been bred; but since, as we have elsewhere 

 shown,* the color of the cocoon may vary in the same species, it is 

 probable that the variation here referred to is not specific. 



•• Xot quite one-half of 1 per cent, produced parasites of various kinds. 



" Limneria pallipes Provanchcr. — In addition to the two Hymenopter- 

 ous parasites treated of, a third one has been very numerous, and has 

 done much good in reducing the numbers of caterpillars. This, an Ieh- 

 neumonid and a much larger insect, does not form an exposed cocoon 

 like that of the other parasites described. Yet a little attention will 

 soon reveal large numbers of them. Upon the trunks of various trees, 

 but chiefly upon those of the poplars and sugar maples, small colonies 

 of caterpillars, varying in numbers from four to twelve, could be ob- 

 served, which did not show any sign of life. When removed from the 

 tree they appeared contracted, all of the same size, and pale or almost 

 white. A closer inspection would reveal the fact that the posterior 

 portion of the caterpillar had shrunken away to almost nothing, whilst 

 the rest was somewhat inflated and covered with an unchanged but 

 bleached skin, retaining all the hairs in their normal position. Opening 

 one of these inflated skins, a long, cylindrical, brown cocoon would be ex- 

 posed ; this is the cocoon of the Limneria under consideration. As num- 

 bers of such inflated skins would always occur together, it was clearly seen 

 that the same parent Limneria had oviposited ia all of them. Most of the 

 cocoons were found in depressions of the rough bark or other protected 

 places. Single ones were but rarely met with. The Hyphantria larva in 

 dying had very securely fastened all its legs into the crevices of the bark, 

 so that neither wind or rain could easily dislodge them. Only half-grown 

 caterpillars had thus been killed. Many of these inflated skins showed 

 in the early part of October a large hole of exit in their posterior and 

 dorsal ends, from which the ichneumons had escaped. Trying to obtain 

 winged specimens of this parasite one hundred and forty of these co- 

 coons — and only such as were not perforated in any way — were collected 

 and put in a glass jar. Only a single female was produced from all up 

 to time of writing, whilst very large numbers ot' secondary paras 

 issued from October 11 till the 20th of November, and doubtless others 

 will appear during the spring of 18S7, because some oi' these inflated 

 skins show as yet no holes of exit. 



"Taehina sp. (Fig. 26). — The parasites of 11. cunea described so far 

 all belong to the order Hynienoptera. which furnishes the greatest nuni- 



* Notes on North American Miorogasters, p. T (author's edition). 



