7° 

 say; the specimens I have been able to investigate were in dried condition and did not show 



this detail. 



In the specimens of the variety robiisia, the chitinous membrane is developed very 

 strono-lv along the line separating the capitulum from the peduncle: here it forms a cushion-like 

 swellino- over the basal portions of the valves of the lower whorl. The ver\- long peduncle 

 seems also a remarkable characteristic of this variety. All the known species of Scalpellum have 

 rather short peduncles, the longest I have seen being that of Sc. Darwini mihi, which measures 

 2/^rds of the length of the capitulum. Of the three specimens of the variety robusta, one is 

 considerably smaller, most probably not full-grown (this specimen is figured PI. VI, fig. 3; its 

 capitulum is 28,6 mm. long, its peduncle 20 mm.); the other two seem to be full-grown: 



the length of the capitulum is 48 mm. in the one and 52,5 mm. in the other specimen. 



the length of the peduncle is 92 mm. in the one and 73 mm. in the other specimen. 



PiLSBRY and Fischer both give figures of this species representing specimens at their 

 natural size. Pilsbry says in his description that the peduncle is long and his figure represents 

 it about as long as the capitulum, which is already a fair length for a species of Scalpellum; 

 Fischer's figure shows the peduncle cut off transversely, only slightly longer than half the 

 length of the capitulum; in his description he says nothing about the length of the peduncle. 

 In the two specimens from Nagasaki I was able to investigate, the dimensions of the capitulum 

 and peduncle are as follows: 



the length of the capitulum is 48 mm. in the one and 44 mm. in the other specimen. 



the length of the peduncle is 30 mm. in the one and 14 mm. in the other specimen. 



There are moreover in the variety- robusta ver)- characteristic and very prominent rings 

 on the surface of the peduncle. The rather large scales are placed in rows which project freely 

 over the surface as so many hoojjs and are separated from one another by rather deep circular 

 o-rooves. The specimen of 48 mm. capitulum-length has about 40, the other about 33 such 

 rings on the surface of the capitulum. They are very distinct on the peduncle of the smaller 

 specimen of the variety robusta also, but their number is there only 14. 



There are in the "Siboga" collection three specimens in all which I consider to belong 

 to this variety. The)' were caught at three different Stations, viz. 



Stat. 5. ;March 10, 1S99. Lat. 7°46S., Long. ii4°3o'.5E. Depth 330 m. Bottom: mud. 



Large specimen, figured PL VL fig- 2. 

 Stat. 251. December 8, 1899. Lat. 5°28'.4S., Long. i32°o'.2E. Depth 204 m. Bottom: hard 



coral sand. Large specimen attached to a piece of shell, which bears numerous 



tubes of tubicolous Annelids on its surface. 

 Stat. 254. December 10, 1899. Lat. S°4o' S., Long. 132° 26' E. Depth 310 m. Bottom: fine, 



grey mud. Small specimen, figured PL YI, fig. 3. 



b. Scalpclhiin Stearnsi^ var. gcniina. 



The variety gemina has imperfectly calcified valves. In the Japan species and in the 

 variety robusta the valves with the exception of the marginal portions are perfectly calcified, 

 the calcified portion having the shape of the valve itself. In the variety gemina the calcified 



70 



