A TAXONOMIC STUDY ON PYROSOMA — METCALF AND HOPKINS. 247 



form we are describing is a subspecies of P. atlanticum. The test 

 walls of the colony are thick (about 6 mm.) and very rigid even in 

 formalin specimens. The colonial aperture is a very small, round 

 pore. 



There are some points of taxonomic interest in connection with 

 the zooids. Of these structural modifications there should be men- 

 tioned the shortening of the oral or prebranchial chamber — except in 

 those zooids which open onto test processes; the displacement of the 

 cloaca dorsally ; a peculiar disposition of the digestive organs, giving a 

 spheroidal appearance to the stomach, and a rotation of the esophagus 

 so that it opens almost from the dorsal side of the pharynx. There are 

 about 30 to 32 stigmatal rows, and 15 to 16 (occasionally 17) longitu- 

 dinal bars in each of the branchial lamellae. The number of dorsal 

 languets is pretty constantly 8 to 9. The length of the zooid varies 

 with the colony from 4.5 mm. to 6 mm., the height from 2 mm. to 

 3 mm. (averaging 2.3 mm.). The cloaca is usually three-fourths as 

 long as the branchial sac, but it may be much shorter than this, 

 especially if the colony is small. It is rather uniformly 2 mm. to 

 2.4 mm., long in zooids from the larger colonies. The cloacal muscle 

 lies at about its middle. 



In this form the gonads are quite prominent, hence easy to observe. 

 The ovary is a large hemispherical compartment, which, when full- 

 sized, bulges out beyond the ventral contour of the body. In its 

 cavity the single, large egg develops until it is set free by the ruptur- 

 ing of the inner cloacal epithelium. The embryo, thus loosed, con- 

 tinues its metamorphosis for a time in the cloaca of the zooid. There 

 are about 14-17 lobes in the testis, which hold together as a fairly 

 compact or rounded-hemispherical body. If a study is made of a 

 small colony (the smallest one referred to above was used) it 

 will be found that practically all the zooids are protandrous; but in 

 the largest colonies the reverse condition obtains; that is, with 

 hardly an exception in the developing zooids, the egg comes to 

 maturity much before the testis has reached its definitive size (fig. 

 34). This seems to be in keeping with Neumann's hypothesis that 

 it is the rapid budding within the colony which delays the develop- 

 ment of the egg and results in a condition of protandry. In this 

 particular variety, it is probable that the colony reaches a definitive 

 size which corresponds roughly to that of our largest specimens, and 

 probably in consequence of this, the rate of budding is somewhat 

 slackened in older specimens and the primitive condition of protogyny 

 is allowed to establish itself. 



Distribution. As mentioned at the head of this description, only 

 six specimens of this form are known. These were all collected in 

 the north Pacific Ocean, somewhere between the Hawaiian Islands 

 and California. Carnegie Magnetic Expedition, between Hawaii 



