38 University of California Publications. [Zoology. 



gonophores were always present in various stages of development 

 ranging from the earliest appearance of buds to medusoids with 

 long manubrium and pulsating bell. This pulsation seemed to 

 indicate that the medusa was ready for liberation and was attempt- 

 ing to free itself. Yet these apparent struggles for liberty have 

 continued, in aquaria, for more than ten days, without success. 

 The bell finally shriveled away, leaving entirely naked the long 

 mouthless manubrium. I have never seen a medusa detach itself 

 nor were they seen to my knowledge in the tow taken daily in 

 San Pedro Harbor during the latter part of May, and the entire 

 months of June and July, 1901, although I have seen eggs on 

 the manubrium of different attached medusa? during the same 

 mouths.* The individuals examined in December had no sexual 

 cells, which makes it probable that the breeding season is limited 

 to the warmer months of the year. 



The perisarc surrounding the lower part of the stem is thin 

 and flexible ; this invested portion of the stem is more transparent 

 than the rest, and its ectoderm contains very few scattered nettle 

 cells, compared with the many that appear at once as soon as the 

 distal edge of the perisarc is passed and cover the rest of the 

 stem. 



C. palma inhabits sand and mud flats between tides, often 

 thickly covering patches many square yards in extent. The 

 filamentous rootlets by which it is anchored, arise as outpocket- 

 ings on the proximal coenosarcal canals, under the perisarc. 

 These small processes, or frustules ( Allman) , may occur regularly 

 in pairs on each canal, or they may be more or less scattered or 

 alternately arranged. There are usually not more than eight 

 pairs on each canal . rather closely associated. The proximal < >nes are 

 longest. Bach process in elongating grows downward for some 

 distance, closely applied to the stem in the manner of a stolon as 

 though responding to a thigmotactic stimulus. The enlarged 



•Since this paper was sent to press eggs of C'orymorpha have been hud in the 

 the laboratory (in May, June, and July, 1902). They drop from the manubrium of 

 the attached medusa, and stick by their adhesive coat to whatever they first touch. 

 There is no free swimming larva. Often the young are clustered on the root fila- 

 ments of old hydroids. The new species of Tubularia recently described by 

 Hargitt (Am. Nat., July. 1902) is undoubtedly based on such clusters of young 

 individuals of C'orymorpha. Buds of the peduncles which support the medusoids 

 appear very early. I shall describe the development more fully in another paper. 



