32 ALLAN HANCOCK PACIFIC EXPEDITIONS VOL. 14 



Robertson's description and figure are inconclusive; she may have had 

 this species or Antropora tincta (Hastings), which is much more abun- 

 dant on the southern California coast, or both of them. O'Donoghue's 

 record from British Columbia is probably correct for his variety triangu- 

 lata, but the varieties paucispina and multispina doubtless refer to some 

 other species, probably Electra crustulenta (Pallas). 



This cosmopolitan species has been somewhat doubtfully recorded 

 from Alaska to southern California, where Robertson states that it is 

 "quite abundant." 



It did not appear in the Hancock dredgings, but the writer has ob- 

 served it in shore collections at Monterey Bay, Newport Harbor, and 

 La Jolla, California. Mr. R. J. Menzies collected it at 5 fms in To- 

 males Bay, California, rather common, one specimen covering more than 

 8 square inches of the inside of a clam shell. Apparently it is not an abun- 

 dant species and has been taken only in shallow water. 



Genus DESMAGYSTIS new genus 



Zoarium encrusting. Zooecia thin walled ; an extensive gymnocyst 

 which is strengthened by a median carina and lateral transverse ribs. 

 Opesia broad, its side walls extended laterally above the bases of adjoin- 

 ing zooecia. Cryptocyst apparently wanting. A median, sessile, transverse 

 avicularium on the distal border of the gymnocyst. No ooecia ; no spines ; 

 no dietellae. Genotype, Membranipora sandalia Robertson. 



Where to place Robertson's M. sandalia has puzzled me greatly. It 

 cannot remain in Membranipora since that genus, as now defined, has a 

 very limited gymnocyst and no avicularia; Electra is suggested by the 

 very extensive gymnocyst, but the presence of an avicularium and the 

 total absence of spines apparently exclude sandalia from the genus ; the 

 absence of any form of ovicell seems to limit it to the simpler Membrani- 

 poridae, but it cannot be accepted in any modern genus of that group. 

 The Membraniporidae {sens sir.) are in so much need of careful, de- 

 tailed revision, that it is with extreme hesitation that I inject another 

 generic description into this difficult group. 



Desmacystis sandalia (Robertson), 1900 

 Plate 3, fig. 1 



Membranipora sandalia Robertson, 1900:324; 1908:264. 



Zoarium encrusting, but not closely attached, forming fan-shaped 

 colonies of a rather rough appearance. The zooecia are large, elongate, 

 narrowed on the proximal half, length 0.85 (0.65 to 1.00) mm, width 



