No. 434.] NOTES ON THE UNIONIDsE. 



only for the animals living in this latitude. But being coinci- 

 dent to a considerable degree with anatomical characters, it 

 certainly is significant, and seems to have some bearing on the 

 phylogenetic origin of these groups, as stated elsewhere. It 

 appears to point to different climatic conditions under which the 

 several groups developed and differentiated from their ancestors. 



Reviewing the points discussed above, we find the principal 

 natural divisions as follows : 



1. The forms with perfect hinges, typically and generally 

 Their embryonic shells are not pointed in the middle of the ven- 

 tral margin, and able to close. 



2. The forms more or less deficient in, or lacking, hinges, 

 typically and generally. Their embryonic valves are subtrian- 

 gular, pointed on the ventral side, and each bearing a large spine. 



Each of these main groups seems to contain the forms nearest 

 related. Any other arrangement would separate naturally allied 

 forms and unite incongruous ones. 



There may be one exception, however, as pointed out above, 

 and that is the genus Proptera. In the members of the genus 

 there are a number of features at variance with those of group 

 1, i. e., the thin shells; the dorsal wings, anterior and posterior 

 to the beaks ; the gaping in front and behind, probably the 

 umbonal sculpture; the slight, and in some species, even 

 deficient hinges. Adding to these the very different form of 

 the embryos, we have a collection of characteristics which place 

 these molluscs not only in a generic rank of equal standing with 

 that of Lampsilis plus its nearly related genera, but as a group 

 by itself. In my opinion they rather represent a third main 

 division between those of Lampsilis-Unio and Alasmidonta- 

 Anodonta. On the other hand they have one characteristic in 



