v-TKltoll.KA Of NOBIS I'Milh IND IDJACEltf WATBR8 PISHKB 7 5 



amount of deflation at the time >>f death. Aside from thee ■ factors there tual 



individual difference in raj breadth. The breadth of ray al widest pa ^ ri t ■» 



K t 5 to 6 nine-, in the largest examples; and i to G 6 times in the medium-sized 

 oens; in small examples there is nol so much variability < >f course in the uln>\ <• 

 comp there is allowance to be made for the increase of raj length with 



There are no two specimens alike in respect to the distribution, number, and Bias 

 of the abactinal tubercles, while their form ranges from that <>f an elongate cone to 

 tliiit of a short acorn, or at the tip of ray tu a depressed spheroid, and the tip may be 

 fairly sharp, to rounded. Not < > r 1 1 > is then- variation in number between Bpecimi 

 but between the rays "f the same example. The tubercles are perfectly solitary, 

 Burrounded only cm the plate by a dense felting of small minor pedicellariae, or then' 

 may be I, -'. 3, i. "i ory tubercles of various, generally smaller, sizes ari 



from the same plate Tin re is scarcely ever more than two largt Bubequal tubercles 



arising from the same plate. Two extremes, that of arm ate tnhereles anil that of 



mostly solitary spines, are encountered at the same station, 3518, ami at this station 

 also i- a specimen with the disk bristling with spines (pi 39, fig 2 and others without 



a Millie spine Aboul the same extremes are found at Station 3251, at which w.r<- 



taken the largest specimens mostlj with few or no enlarged tubercles on disk). 

 (PI. 39, fig. 1.) 



A curious variation is found in a specimen from the Arctic Ocean, north of 

 Bering Strait-* (steamer ( brurin, No 15819); R 96 mm., r 19 mm. The dorsal surface 

 i- typical, hut the actinal -pine- are lacking opposite the first 8 to \'J inferomarginals. 

 From tin- point on there i- a single series which extends aearlj to the tip of the ray. 

 There is •■<■ single series of verj short Buperomarginal Bpines and proximally a second 



-eric- of shorter tnhereles. The inferolnalLritials are abruptly much lar.'er 

 and dominate the amhitu- The straight pedieellal iae of the dorsolateral and mtir- 



ginal regions are large, with broad, toothed extremities Thi- is not a variety char- 

 acteristic "f the Arctic Ocean for another specimen collected l>\ the Cortrin, 

 No 15819, Arctic Ocean, north of Bering Strait, has much shorter rays, a conspicuous 

 intermarginal channel, and a relatively broad area between the inferomarginal and 

 adambulacral plates occupied l>\ a series of -tout actinal Bpines 



- cimena not typical and variously divt -ward acervOta were taken at 



the following Stations, no two of these Bpecimens being closely similar: Station I 



one intermediate; 3252, one; 3253, on< 77, three; 3506, oi two 



intermediate, the proximal part of the ray of example being <u; rnitn and tin- 

 tip, pclyihda; 1795, one; from this station a typical pdythda. Arctic Ocean, north 

 of Bering Strait, one PI 38, fig 



Young.- From station 3254 along with adults ami undoubted young is a tiny 

 specimen having -ix ray- and R I 5 mm. Tins is chiefly interesting in having the 

 abactinal surface covered with thin Hat subcircular, overlapping plates from bod 



which arises an erect spinelet ; no papulae; tw ties of tube feet ; mouth plate- 1 



like those of the SohiMetidae. with two actino-tomial -pmelet- ami at the outer end 



of the plate a larger Buboral; one pair of adambulacrala touching behind the mouth 

 plate-: one large adambulacral spinelet; actinostome big; a brief series of marginal 



-pinelet-in each row Thi- -pecimen is of OOUISS quite unlike the adult and re-emhle- 



one of the Asterinidae mote than an Asteriidae 



