BEARING OF LAEViE ON CLASSIFICATION. 37 



the lar«::er. The hypostomal sclerites have a hump or swellint^ 

 beneath; the great hooks have hardly any spur above at base. In 

 some of the <^enera, CalUphora and ProtocaUiphora, onW part of the 

 upper hmb of the lateral plates is strono;ly cliitinized. 



In Ilomalomyia (fig. 127) there is an approach to the form of 

 Mnsca, the lateral plates only slightly indented, the lower limb the 

 larger, and a basal spur above on the great hooks. In the upper 

 anterior part of the lateral plates, where they unite, is a number of 

 perforations, some round, some elongate; these are not seen in the 

 other families examined, but appear in Drosophila (fig. 133). 



BEARING OF LARV.E ON CLASSIFICATION. 



The value of larval characters in classification ^\■ill always be 

 variously estimated by different entomologists, and the \vi'iter is far 

 from claiming that any group should be dehmited b}^ larval char- 

 acters. But in view of the ch versify of opinion among dipterists 

 as to the divisions of the old family Muscidse, the bearing of these 

 larval structures may be of interest. Some authors have taken 

 Stomoxys and Glossina from the Muscidae and put them in a separate 

 family, the Stomoxyidae. From the standpoint of the larvae there is 

 no warrant for this separation, Stomoxys being much nearer in struc- 

 ture to Musca than is either to the Calliphorinae. Looking at these 

 gi'oups from the structure of the larvae, one will notice that both the 

 Muscidae and the Anthomyiidae possess two styles of larval structure, 

 and it is very difficult to see why these striking differences in the 

 larvae should not find some corresponding difference in the flies upon 

 wliich to found a better classification. The larva of Homalomyia 

 differs so greatly from that of Anthomyia that one can not but tliink 

 that this difference should be reflected in the adult. Like\\dse when 

 one considers the peculiarities of the larvae of Musca, Stomoxys, 

 Lyperosia, and Pseudopyrellia, difl'ering from other muscids by impor- 

 tant characters at each end of the body, one can hardly believe that 

 there is not some structure of the flies to distinguish them as a group. 



The classifications of the Muscidae by Pandelle and by Girschner 

 agree much better with the larvae than the classifications seen in the 

 catalogues. These authors have put the (^illiphorinae remote from 

 the true Muscidae and near the Sarcophagidae; Pandelle also has a 

 distinct group in the Anthomyiidae for Ilomalomyia, and the true 

 Muscidae are included in his Anthomyiares. However, the larvae of 

 the true Muscidae are so different from Anthomyia that one would 

 suppose the flies should have a group at least equal in value to that 

 of the Anthomjaidae or Tachinid». This would iutUcate three 

 famiUes, Muscidae in the restricted sense — Tachinidae, to include 

 Cafliphorinae and Sarcojihagiila? ; and the Anthomyiidje, to include 

 Musdna and Homalomyia. the latter to have subfamily rank. 



