18S1.] PROF. F. J. BELL ON THE GENUS ASTERIAS. 509 



Series y. Outer Skerries, Scotland. Breadth of ray. 



i. .. . B=l0-5, r=3-0, or E=3-5r. 3-5 



ii R=100, r = 30, or B^3-3r. 35 



iii 11= 9-5, r=3-0, or E=3-I6r. 45 



iv B= 9-0, r=30, or B=30r. 3-7 



V B= 7-5, r=2-5, or B = 30r. S'O 



Arranging these numbers in order, we find for A. rubens 

 3-83, 3-6, 3-5, 325, 3-1 6, 3-16, 3-15; 

 and for A. hispida 



3-5, 3-4, 3-4, 3-3, 3-16, 3-0, 2-619, 2-437, 2-419, 2-0. 



For A. rubens, then, the proportion of fl to r does not fall below 

 3-15: and the specimen in which that obtains is comparatively 

 young, and has R only equal to 8'2 millim. ; in A. hispida it may 

 fall as low as 2, and that in a specimen of moderate size, or with a 

 long radius of 12 millim. 



If we take only sei'ies a of A. hispida we find a marked shortness 

 of ray ; but, unfortunately, that series is so far aberrant that the 

 smallest specimens have R proportionally the longest. If, on the 

 other hand, we take series y (and that is the series which Mr. Nor- 

 man has presented to the British Museum), we find a set of propor- 

 tions which can hardly be said to compare unfairly with series /3 of 

 A. rubens. 



If now we take a series of older specimens of A. rubens, we may 

 be led to certain conclusions ; the set taken, though not in any way 

 specially selected, is very favourable : — 



i B=Gd, r=13-5, or i2=4-81 r. 



ii B = &Q, r=13-0, or 72=4-61 r. 



iii 72=56, r=13-0, or i? = 4-G0r. 



iv ^=46, r=120, or i2=3-83r. 



We are therefore, as it seems to me, led to the conclusion that 

 for A. rubens there is a gradual increase, during growth, in the 

 length of B as compared with r. 



In some cases, as with series y, sets of specimens of A. hispida 

 may well fall into the series ; but, on the other hand, they may (as 

 does series a) exhilnt so great a variation as to lead to the sus- 

 picion that they must have been subjected to more or less abnormal 

 conditions. And, finally, we cannot with safety appeal to the 

 shorter rays of A. hispida as a definite and constant mark of specific 

 differentiation. 



The next specific character given by Forbes is "rays rounded;" 

 but the same term is applied to A. rubens ; then we have " spinous " 

 for A. hispida, and " very spiny " for A. rubens. 



The last character, and one which is regarded as being a good 

 mark of distinction in this genus, is the form of the "ambulacral 

 avenues;" they are lanceolate in A. rubens, and "elongate-ovate" 

 in A. hispida. As it is obvious that the form of the ambulacra will 

 depend not a little on the width of the rays at their base, I proceed 



