210 Mr. II. Scotl ov 



abdomen consists largely of collapsible connexival membrane. 

 Hence -with complete dessication tlie segments become so 

 sbrunk, distoi'ted, and telescoped one under the other that 

 specific characters may be made almost entirely unrecogniz- 

 able. The same statements apply in a less degree to the ($ . 

 It is practically impossible to determine species from some 

 of the earlier descriptions, which have been rendered incom- 

 plete or false by the dried and shrivelled condition of the 

 material from which they were made. jSIucli error and. 

 confusion is likely to arise in consequence of this. By a 

 thorough soaking in water old dried specimens can some- 

 times be induced to reassume in some measure their proper 

 form, and they can then be gradually transferred to weaker, 

 and afterwards to stronger, alcohol. 



In studying Nycteribiidse — even those properly preserved 

 in alcohol — very great allowance must be made for the 

 changes in appearance caused by the varying condition of 

 the abdomen, particularly in the ? sex. Descriptions and 

 figures will differ greatly according as they are taken from 

 $ $ in an advanced stage of gestation or from specimens 

 •with empty shrunken bodies. For the same reason two ? ? 

 of the same species may look so different from one another 

 that careful study of the parts of the abdomen is required to 

 establish their specific identity. 



For the same cause it is sometimes impossible to include 

 in a single camera-lucida drawing all parts of one aspect 

 of the ahdomen. For example, in a dorsal view the basal 

 segments may, owing to curvature, be almost perpendiciilar 

 to the field of vision. In such cases it may be necessary to 

 make two outline drawings of tl.e specimen tilted at different 

 angles, and to compound them into a single figure showing 

 all the segments. Allowance must be made for this in 

 using the figures on the Plates attached to this paper. 



Fault may be found with the great length of some of the 

 writer's descriptions and with the omission of short dia- 

 gnoses ; but it seems at present impossible to ayoid making 

 very long and detailed descriptions of these insects. Such 

 important characters are presented by the abdomen that it 

 appears necessary to describe in detail the form of each 

 dorsal and. ventral segment in both sexes. Much trouble 

 arises from the inadequateness of earlier descriptions; a 

 number of species have been placed in wrong genera, and. 

 this cannot always be discerned from their descriptions 

 alone. jNIoreover, the writer is uncertain whether the 

 present system of genera is satisfactory. AYhen the generic 

 position of the species is more settled, and those of the same 



