THE AMERICAN XATUKALIST. [X oi.. XXXVIII. 



uals at the Tortug-as, for a surprising- number of characteristic 

 species appear to be confined to this region alone. Moreover, 

 at the Tortug:as, we have a small land mass surrounded by pure 

 deep ocean water, while at the Bahamas we find a vast area of 

 shallow flats covered mainly by coralline mud, forming veritable 

 submarine deserts covered only with a sparce growth of coralline 

 algai and a few scattered clusters of coral reefs and gorgonians. 

 The water ov^er these shallow banks is almost as deficient in life 

 as is that of the desert bottom itself, very few Sagittal, Salpje 

 or Crustacea being found, and among the Medusae only repre- 

 sentatives of the Gonionemidae and Lymnorea are abundant. 

 Indeed, the water of these shallow banks is usually charged 

 with a flocculent mass of silt which adheres to pelagic animals, 

 and appears to be rapidly fatal to them. The deep water regions 

 of the Bahamas, however, such as the tongue of the ocean, or 

 the water to the eastward of Great Abaco Island, are rich in 

 pelagic animals which appear to be free from silt and in good 

 condition. This water of the deep areas must, however, be 

 often driven over the shallow banks by the winds, and its life be 

 thus destroyed. 



It is found that Olindias is closely related to Gonionemus, for 

 in its ontogeny it passes through a stage in which all of the 

 tentacles arise from the side of the bell, and are sucker-bearing 

 as m Gonionemus. EucJieilota paradoxica is the only Leptome- 

 dusa known which produces an asexual generation of Medusae 

 by a direct process of budding. These daughter Medusae are 

 derived from both ectoderm and entoderm of the gonad of the 

 parent, as is the case in the Sarsiad^e and in Hydroids. On the 

 other hand, the Medusae buds found upon the manubrium of 

 Boiigainvillia uzobe are developed entirely within the ectoderm, 

 the entoderm remaining inert and passive during the growth of 

 the bud, its limiting membrane being unbroken, and no connec- 

 tion ever being established between the gastro-vascular cavities 

 of the bud and the parent. In B. niobe, however, the ectoderm 

 of the manubrium is of considerable thickness, affording abun- 

 dant material for the formation of the bud. It is possible, 

 therefore, that this peculiar method of formation of medusa 

 buds from the ectoderm, which has been observed in B. niobe by 



