PENNARIA. 185 



cells arranged irregularly, not in definite clusters as they are in E. dumortii. 

 But in a series from New England, in the collection of the Museum of 

 Comparative Zoology, I have found them forming quite as definite swell- 

 ings along the outer sides of the tentacles as they do in the present 

 specimens (PL 38, fig. 12). It appears, then, that this character is a 

 variable one, and therefore not of much taxonomic importance. At any 

 rate, specimens both from the Atlantic and from the Pacific exhibit the 

 same features in this respect. Hargitt (: 05'') has described ocelli in this 

 form ; but in neither Atlantic nor Pacific specimens have I been able to 

 find anything more definite in the way of light perceptive organs than the 

 brilliant pigmentation in the centres of the basal bulbs of the tentacles. 

 E. dumortii is said to have no ocelli (Hartlaub, : 07). 



Color. — I was unable to examine the specimens in life. After preser- 

 vation in formalin the manubrium is ochre yellow, the tentacle bulbs 

 pale yellowish. Preserved specimens from the Atlantic have about the 

 same color, but in life they are very brilliant, the tentacular bulbs being 

 yellow with bright red centres. 



Pennaria Goldfuss, 1820. 



Globicep3 Ayers, 1852. 



Codonidae with all four tentacles entirely rudimentary ; manubrium 

 short, not extending beyond bell-opening; no ocelli. Hydroid stage, 

 genus Pennaria Goldfuss. 



In this genus specific characters are to be sought in the hydroids rather 

 than in the very rudimentary medusae of those species in which they 

 are developed, and both hydroids and medusae, at least of P. tiarella, are 

 very variable. The present collection contains two species of Pennaria, 

 from widely separated localities, each represented by but a single speci- 

 men. One of these agrees so well with the account of P. vitrea Agassiz 

 and Mayer ('99), from the Fiji Islands, that I refer it to that species. The 

 other most nearly resembles P. tiarella, yet differs from it in so many 

 respects that it probably belongs to a different species. As yet, so far as 

 I can learn, no species of this genus has been recorded from the west coast 

 of America although two species, P. rosea and P. adamsia, have been de- 

 scribed from Australia by Von Lendenfeld ('84), so that there is no clue, 

 except that afforded by the medusa itself, to its identity, and since in this 



