310 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



and then, approximately in the range given by Carpenter, that is from the tenth to 

 the twentieth brachials, they move nearer the base, to the height of the third to fifth 

 or third and fourth segments. The third to fifth, or third and fourth, or fourth and 

 fifth segments are then flat and laterally broadened. This broadening, which Hartlaub 

 frequently pointed out in connection with other varieties, is purely an accompaniment 

 of the swelling of the gonads and is similarly developed in other specimens of brevipinna. 

 It is usually lacking in smaller (younger) individuals since in them the gonads contain 

 no ripe sexual cells. Hartlaub expresses his surprise that Carpenter, to whom this 

 relationship between the ripeness of the gonads and the expansion of the pinnule 

 segments covering them must have been known, should have employed such an un- 

 stable character for the separation of the two species. He said that he must strongly 

 oppose the supposition that the pinnule segments overlying the gonads in Carpenter's 

 specimens of pourtalesi are significantly and constantly stouter than those in typical 

 specimens of brevipinna — for example, in his var. eUgans. The assumption that rel- 

 atively slender genital pinnules with only slightly broadened segments are character- 

 istic of brevipinna is untenable. So Hartlaub saw no reason for accepting Carpenter's 

 new species. 



As already mentioned, the two specimens determined by Carpenter as pourtal&si 

 show rather wide differences. The smaller specimen shows spiny ornamentation, 

 and therefore will be described under the varieties of the Spinosa type. The larger 

 specimen, because of its developed ornamentation is naturally associated with the 

 varieties of the Tuberosa type. Its characters are as follows. 



The centrodorsal is subhemispherical with a scarcely noticeable dorsal flattening. 

 The bare uneven dorsal pole is small with an inconspicuous central cavity. It bears 

 a few rather sharp tubercles, which are also found between the cirri and on the slightly 

 developed interradial processes. 



The cirri are XXXVI ; none of them are preserved entire. According to Carpenter, 

 who saw the specimen when it was in a better state of preservation, there were 15-20 

 cirrus segments. The distal segments, according to Carpenter, are smooth, a state- 

 ment which he applies also to brevipinna, but according to Hartlaub's observations 

 does not apply throughout; rather in brevipinna the distal edges of the segments in the 

 outer half of the cirri are more or less produced, and the penultimate bears an opposing 

 spine. 



The radials are wholly concealed. The IBr[ are flat and bowed into a kind of 

 scallop with the convexity proximal. The IBr 2 (axillaries) are relatively flat and 

 approximately rhombic. The form of the elements of the IBr series recalls forcibly 

 that in the var. tuberosa. All 10 IIBr scries are 2, and the 12 HIBr series also are all 2. 



There are 32 laterally compressed arms of which 8 arise from IIBr axillaries and 

 the others from IIIBr axillaries. In the region of the proximal brachials there is to 

 be noticed a double change in form which is best observed on the basis of the direction 

 of the articulations. The articulations of the first three or four brachials are transverse, 

 of the following three or four more or less diagonal, of the next two or tliree transverse, 

 then strongly diagonal to the arm tips. This feature is much more strongly brought 

 out in the specimen referred to the Spinosa type (var. coronata). 



The ornamentation of the division series, as already noted, agrees naturally with 

 thai of var. tuberosa, appearing here in reduced form. Also here the basic element of 



