﻿31 



with its bizarre shapctl head, and Mi\i:;atn)l^is, with its wonderful, 

 separate head-keels. 



The third group is distinctly divided into two subgroups, the 

 relationshi]:) between theni being very indistinct and uncertain. 

 The Sikicvia subgroup consists of delicate little creatures, with 

 wings apparently wholly devoted to sound production ; the Zo- 

 raida subgroup also has very small wings with large anal areas, 

 devoted to sound production, but never so greatly reduced as in 

 the Sikiana section. It is among the Zoraida section that the 

 most conspicuous forms of the family are to be found, and from 

 whence the forms generally chosen to illustrate the family are 

 taken. Among the genera of this section Zcui:^uia stands apart 

 as possessing both shoulder keels and subantennal processes and 

 demonstrates the uselessness of the structure of head and thorax 

 for primary divisions. Diostronibus and Comma are the two 

 forms which approach nearest to the Sikiana section. 



The last group is not a natural one ; it consists of Dcrbe and 

 Mysidia, which approach the Zoraida section of group three, the 

 Rhotaiia section, which is somewhat aberrant, and the Nisian 

 genera, which are not true Derbids. There is a touch 

 of irony in the fact that Dcrbc, the type genus of the family, is 

 not truly characteristic of any of the main groups. By discard- 

 ing Xisia and its allies, and placing Dcrbc and Mvsidia in the 

 Zoraida group, the small Rhotaiia group could remain unat- 

 tached. 



Unfortunately there arc certain genera of which the descrip- 

 tions are not com])lete enough to enable me to include them in 

 my tables. Working in a place away from large collections and 

 libraries, one is dependent upon one's friends more fortunately 

 situated, ^fy thanks are due to the United States National Mu- 

 seum for sending me specimens for examination, and to Mr. W. 

 L. Distant of the I'ritish Museum, for specimens, answers to 

 many questions and nmch valual)le information. 



Rohan and Pekalongan are in Java. Telok Ayer. Pontianak 

 and Mowong in I'orneo. and Laloki in Pritish Xew (aiinea. 



When working with small and delicate injects like this family 

 I use a Greenough binocular, as a single hand-lens is not suitable 

 to recognize carinae. etc. 



The measurements are horn front of head to end of pygophor, 

 not including the genital styles, and from base to apex of one 

 tecrmcn. 



