^10 GEOL. AND NAT. HIST. SURVEY OF MINNESOTA. 



collector. This family uadoabbedly is the link coaaectiag the Daph- 

 nidce with the Lynceidce, relationships to which are expressed by 

 Macrothrix, on the one hand, and Lyncodaphnia on the other. 



The rank of this group as a family must be, ol course, a matter 

 largely of opinion. Sars was the first to adopt this view, sustained 

 by certain curious transition forms leading toward Lynceidce. Later 

 -writers seem never to have found these genera and the group was 

 again included with the Daphnidce. The writer ('82), upon the discov- 

 ery of the Lyncodaphnia was forced to regard this gioup as of equiva- 

 lent grade with the above mentioned families and again proposed the 

 family name Lyncodaphnidce. 



The genus Ilyocryptus is a little one side the normal course of the 

 family and seems related to the lynceid genus Leydigiu. 



The waters of the northern United States are very rich in members 

 of this family. 



The aberrant family Bosminidce finds its only connection with other 

 Gladocera through this group by means of the remarkable Macro- 

 thrix {1) 'p^nper; and here it is only vaguely hinted at in the elongated 

 antennules and angled lower margin of shell, as well as the presence of 

 certain bodies near the base of the antennules. It has been affirmed 

 that none of the Lyncodaphnidce have an ephippium, i. e., the saddle- 

 shaped thickening of the shell walls to include and protect the winter 

 eggs; but I have discovered it in the case of Macrothrix tenuicornis 

 Kurz and presume it may occur exceptionally in others. Ivurz says 

 that Ilyocryptus has no moult proper, but this probably refers only to 

 the European I. sordidus. The American species differs from the 

 generic description given by Kurz, and may be different in this 

 respect also. 



In this family the regularity in the disposition of the setse on the 

 antennsB is broken and the fringing of these hairs serves the purpose 

 of specific distinction. The antennules are always long and frequently 

 differ considerably in the sexes. The pigment fleck is always present 

 (Kurz is in error in denying its existence in Lcdhonura). In many 

 forms there is no free posterior margin of the valves, while the lower 

 is generally thickly beset with movable spines. The Lyncodaphnido} 

 will be distinguished from Ceriodaphnia, which they resemble, by their 

 motion, which is a succession of quick bounds, while the broader 

 Ceriodaphnice hobble along as though heavily weighted by the enor- 

 mous mass of eggs with which they are generally laden. The abdomen 

 is usually short and the anus is behind the terminal claws, but in Ily- 

 ocryptus the claws are long and spined at the base. In the American 

 I. spinifer the anal opening is elevated to a i)oint nearly underneath 

 the stylets, and there is a rudimentary anal caicum as in Lynceids. 



