228 ANNALS OF THE CARNEGIE MUSEUM. 
characters which are essential, since they indicate the various ‘‘ideas”’ 
in the specialization within each group, and advance our understanding 
of the phylogenetic progress and the systematic affinities of the Na- 
jades. 
It may not be amiss to point out that it is absolutely impossible to 
recognize this system in the characters of the hard parts, the shells. 
It is true that certain types of shell are characteristic within smaller 
groups, and that there are cases, where we are able to recognize a genus, 
for instance, by the shape of the shell. But if we come to compare 
the subfamilies and families, we find that various types of shell turn 
up in them again and again. This goes so far that certain species 
resemble each other so much externally that they have been confused 
or placed together even by our greatest authorities, while they actually 
may belong to entirely different groups according to the soft parts. 
For this reason I have deliberately omitted to give shell characters for 
the families and subfamilies, for this is simply impossible. 
One character of the shells, however, may be of greater value, and 
this is the beak-sculpture. As will be seen below, I shall use it re- 
peatedly for the definition of genera. But it has been largely misunder- 
stood, and is even now not very clear. Simpson, in distinguishing a 
concentric and a radial beak-sculpture, made a great mistake in uniting 
under the latter two types of sculpture, the radial and the zig-zag, 
while he united the double-looped with the concentric sculpture. 
According to my studies, which, however, are not yet fully satis- 
factory, the following seem to be the real conditions: The original and 
simplest beak-sculpture consists of concentric bars. A few (one to 
two) of them are, when the beaks are well preserved, always present, 
even in zig-zag or radially sculptured beaks. In many forms other 
bars of the same character are added, and no complications are ob- 
served. In other forms the later bars become double-looped. This 
character is generally inaugurated by the fact that the posterior part 
of the simple bar, which lies upon the posterior ridge of the shell, is 
emphasized. It becomes more pronounced, often tuberculiform, and 
is drawn out in the direction of the posterior ridge, toward the lower 
posterior angle of the shell. This produces an angular projection in 
the posterior part of the original bar, which by contrast with the 
anterior part, which does not project, gives the appearance of the bar 
consisting of two parts, or two loops, till we finally come to a beak- 
sculpture which distinctly consists of a double loop, the two parts 
