ATTACHMENT OF THE ABDOMEN TO THE THORAX IN DIPTERA 257 
stuck into the plasticine just above each end of the insect. For drawings 
requiring two or three hours this proved to be a very satisfactory method, 
as the alcohol remains clear for that length of time. 
In each drawing an attempt was made to show all chitinized areas 
clear and all membranes stippled. There are sclerites to be found, 
however, in which it.is difficult to class the integument as either chitinous 
or membranous. Assuming the membranous state to be the more primi- 
tive, increasing amounts of chitin laid down in membrane were represented 
by a decreasing amount of stippling, that is, by placing the dots farther 
and farther apart. All outlines, as well as sutures (using the term in a 
general sense), were shown in full lines, while all endoskeletal parts, 
such as phragmas, and all sutures covered over by other parts, such as 
appendages, were shown in the dot-and-dash line. In most figures the 
amount of development of the phragma between the meso- and the meta- 
tergum also was represented in this manner. Indistinct sutures and 
boundaries of chitinized areas which shade off into membrane, or vice 
versa, were represented by dotted lines. Spiracular openings were shown 
either with a crosshatched interior or with a fringed border. 
Pencil drawings were inked on the coordinate paper and plates were 
made directly from these. Drawings were either enlarged or reduced in 
order that all plates might have one dimension, the length, uniform. 
‘Prints were made on contrast paper and the sclerites were labeled on these 
reproductions. 
MATERIAL 
The material on which this paper is based was all drawn from the col- 
lections in the Department of Entomology at Cornell University. Dis- 
sections have been preserved in alcohol and retained for reference. 
For the convenience of systematists the figures are arranged in the 
order of the families to which each species belongs, as listed in Aldrich’s 
Catalogue of North American Diptera. 
Of the fifty-nine families according to Aldrich, one or more species 
from fifty-five have been studied and figured. In addition Sczara ochro- 
labis and Piophila casei, which are included by Aldrich under the families 
Mycetophilidae and Sepsidae, respectively, are figured and placed in 
the families Sciaridae and Piophilidae. The following families are not 
represented: Acanthomeridae, Apioceridae, Phycodromidae, and Nycteri- 
