128 PROCEEDINGS OF THE ACADEMY OF [Jan., 



gizzard, to be very variable. lie examined the gizzards of 

 Hetizriiia, americani, Archilestes grandis, Leste-s tenuatus, Argia 

 pulla and agrioides, Eri/thagrion salvum, Ischnura Ramburii var. 

 credala, Ana.v Junius, H'irpetogomphus elaps, Pantala hymencea, 

 Pseudoleon superhm, Tramea onusta, Dythemls sterllis, Micrathyria 

 Hagenlf Orthemis ferruginea and D'lplax corrupta. 



Dr. Ris and Dr. Calvert agree generally on the morphology of 

 the forms studied by both. They differ, however, in the phylo- 

 genetic position of the legion Lestes. 



Dr. Calvert does not draw any phylogenetic conclusions from his 

 studies, considering the examination of many more forms neces- 

 sary before this can safely be done. 



Preparation of Present Material. — The present study has been 

 based upon representatives of the sub-families Calopterygince and 

 Agrionince collected from every continent. 



The preparations have largely been made from dried^ specimens. 

 From these the abdomens were cut off at the base of the third 

 segment and soaked in seventy per cent, alcohol until softened. 

 Some of these, even after long soaking, were so brittle that only 

 fragments of the gizzard could be mounted. In cases where, 

 because of its fragmentary condition, the structure of the gizzard 

 is at all doubtful, I have put an interrogation point after the 

 descriptive formula. The remainder of the material studied had 

 been preserved in alcohol or formaline. 



To obtain and prepare the chitinous lining of the gizzard for 

 study, the abdomen was slit open along one of the membranous 

 pleura, the gizzard located, and separated from the rest of the 

 ahmentary canal by a fine pair of scissors. The gizzard was cut 

 open, the muscular coat removed with nesdles, aad the chitiuous 

 lining spread out flat upon a slide. After cleaning, this was 

 mounted in Canada balsam, making a very distinct preparation for 

 study. In nearly every case the dried specimens from which the 

 gizzards have been removed have been again put together, so that 

 their value as ordinary museum material has been little, if at all, 

 impaired. 



The larvae studied belong mainly to the sub-family Agrioninoc. 

 They were collected from the vicinity of Philadelphia and were 

 either fresh or preserved in alcohol. 



* See the. report of the meeting of the Academy of Natural ficieuces of 

 Philadelphia of January 17, 1899, in Science, Vol. IX, p. 183. 



