4 PROCEEDINGS OF THE NATIONAL MUSEUM. Vol. 66. 



based on Eisen's first paper v.oald be aa :ah:bc: cd=7 : 3 : 6 : 4. In 

 his last two papers he writes " a^h slightly larger than c-d.'^) Ven- 

 tral setae are lacking on 21; and those of 20 and 22, related to the 

 prostate gland pores, differ but slightly from those of other somites, 

 being somewhat straighter. The ventral setae of the spermathecal 

 somites are not modified in character, as are those of several other 

 species of the genus. Similar statements are made by Eisen con- 

 cerning D. keyesi. 



The clitellum on the dorsal side includes all of 13-20 and a part 

 of 12. Ventrally it extends from i/^j 13-1/^20, but thins out on a nar- 

 row median strip. A transverse glandular thickening extends across 

 this median area between the oviducal pores. Eisen describes and 

 figures two longitudinal grooves connecting the prostate gland pores 

 of the corresponding sides. No such grooves are present in the speci- 

 mens of the new form, but the absence is probably due to the sexually 

 nonactive condition. Such gi-ooves are present at the time of sexual 

 activity in most known species of the genus. The first dorsal pore 

 is at the anterior margin of 10 in the type, and in 11 of the paratype. 

 The nephridiopores are near the anterior margins of the somites. 

 Those of the most anterior pair are in the second somite and are 

 dorsad of seta line d by a distance at least twice as great as cd^ the 

 distance between the setae of the dorsal bundle. Several of those in 

 the clitellar somites are dorsad of seta line c? by a distance nearly as 

 great as cd^ or even greater. The majority of those of the anterior 25 

 somites are but slightly dorsad of seta line d. The nephridiopores 

 of somites postenor to 25 have not been studied. The spermiducal 

 pores are on the ventral side of 21 ; are scarcely one-fourth of the 

 length of the somite from the anterior margin; and are nearly in 

 seta line h. The prostate gland pores are in close relation to seta h 

 of somites 20 and 22. Eisen describes and figures the spermiducal 

 pores of D. keyesi as opening midway of the length of 21 and midway 

 between the corresponding prostate gland pores. In the new variety 

 the distance from the spermiducal pore to the anterior gland pore 

 of the same side is scarcely one-half as great as that to the posterior 

 pore. The oviducal pores are on the anterior part of 14 and slightly 

 mtsad of seta line a. The spermathecal pores are near the anterior 

 margins of 8 and 9 and nearly in seta line a. 



INTERNAL CHAR-VCTCKS. 



Septa 7/8 and 8/9 are most strongly developed and are about as 

 thick as the body wall. Septa 6/7 and 9/10 are also thickened, but 

 not as much as the two first mentioned. Septum 5/6 is normally 

 developed and extends to the body wall, and so also does an extremely 

 thin but perfectly evident septum 4/5 which is the most anterior 



