LYROPUPA. 235 



three figures sho\v a sinistral shell, which may be taken as 

 the type. This shell is drawn in pi. 19, fig. 4; or at least so 

 we assume, as only this one specimen had the aperture well 

 cleaned when the lot was examined. 



The spire is convexly conic. The ribs are strong: through- 

 out, 17 on the last whorl (Gould probably doubled the number 

 seen on the face to reach the number "ad 20"). There is 

 a rather deep external impression about a half whorl long 

 over the upper palatal, and a shorter one, diverging forward, 

 over the lower palatal. The lower palatal fold is rather long, 

 as seen in the mouth. It becomes stronger towards the inner 

 cud, which penetrates slightly beyond the upper palatal fold. 

 It is not hooked downward at the inner end, and not inter- 

 rupted, though in some specimens of the lot. there is a sub- 

 median depression of the crest of the fold. Fig. 5 represents 

 the palatal region of an example of the Gould series. Some 

 others have the lower palatal fold somewhat shorter and 

 lower anteriorly (fig. 6) . 



Length 2.7, diam. 1.6 mm.; 5 1 /', whorls (type). 



Length 2.4, diam. 1.6 mm. (another of type lot). 



The contour varies, as the measurements indicate, the shorter 

 examples being shaped like pi. 19, fig. 1, which represents a 

 paratype of magdalenm*. 



With this typical, strongly ribbed form of lyr<it<t there are 

 other specimens, probably collected from a different colony or 

 locality, in which the ribs are weak and irregularly developed 

 or partly wanting on the last whorl or two (pi. 19, figs. 8, 9). 

 These may be called L. lyra-ta- form gouldi. 



Pease's locality for this species, "Oahu, " was based doubt- 

 less on correctly named specimens. Boettger's description of 

 lyrata was founded 011 specimens of another species from 

 Hawaii. These are undoubtedly the same specimens used by 

 Pfeiffer for his description. Ancey's description of I y rat a 

 was based on specimens of an unnamed species for which the 

 name L. anceyano is proposed. 



Pupa magdalenae Ancey (type no. 18744 Bishop Mus., para- 

 types 18745 B. Mus. and 119455 A. N. S. P.) is not separable 



