August, 1902] 



rSYCI/E. 



•All 



cle. While the most striking forms of Our caddis-fly, however, is apparently 



prepupae hitherto described {Agrioytpiis^ worthy to be ranked as an illustration of 



etc.) have been pointed out as cases of a third type. Besides the three normal 



hypermetamorphosis, it is to be noted developmental stages, it takes on in the 



that they are all forms strictly interme- prepupal period characters which do not 



diate between the larva and the pupa, belong to either larva or pupa, and are 



and all exceedingly transitory, gradually not intermediate between these stages, 



but quickly assumed and (juickly lost The possibility that my specimens did 



again, and not truly hypermetamorphic, not all belong to the same species, and 



in the same sense as are the two types that these peculiar appendages of the 



mentioned above. prepupa may have belonged to the larva 



