DR. P. H. CARPENTER ON THE CIRRI OF CERTAIN COMATUL.E. 479 



This, which may be called the square-jointed type, furnishes the greater number of 

 the cirri in the Scotch specimens. A very characteristic example consisting of 51 joints, 

 but only 47 mm. long, is shown in fig. 29, and another variety with the terminal joints 

 longer in proportion to the width than those just below them in fig. 28; while fig. 21 

 shows another form with shorter joints and two opposing spines. The three youngest 

 cirri of this type that I have been able to find are shown in figs. 20, 22, and 2G. The 

 first two of these are from quite young individuals ; but the third is from a much older 

 one. The opposing spine and the carination of the later joints which is prominent in 

 the two former is less evident in the larger cirrus, which has rather the appearance of 

 having been developed on the rudi mental plan (compare fig. 31) ; while the two small 

 cirri of the young individuals clearly belong to the small mature type of Dr. Carpenter. 

 The smaller of the two is distinctly shorter, though with wider terminal joints than a 

 rudimental cirrus of type B, from another young individual at the same locality. 



Type D. Short-jointed. — Besides the square-jointed form of cirrus just described, which 

 is the prevailing one in the Scotch variety of Antedon phalangium, several others may 

 be found with the later joints distinctly wider than long. The most extreme form of 

 this short-jointed type which I have seen is shown in fig. 33. Figs. 30-32 seem to 

 represent earlier stages of its development, though I am by no means clear that the last 

 one, if not the other two as well, may not belong to the square-jointed type shown in 

 fig. 29. They are interesting, however, as furnishing an example of the rudimental 

 process of development which we have also studied in the long-jointed Mediterra 

 variety (figs. 1-6) ; and a comparison of the two brings out very clearly the difference 

 between the latter and the short-jointed northern form in their immature condition. 



Thus in a young cirrus from the Minch shown in fig. 31 there is a distinct terminal 

 claw, and the later joints have almost assumed their final shape; whereas in the much 

 longer but very undeveloped cirrus of the southern type (fig. 6) there is only the merest 

 trace of a terminal claw, and the later joints are still in a discoidal condition. In a later 

 stage, however (fig. 4), they have lengthened out and begun to look more like those of 

 the full-grown cirrus (fig. 7); while in the corresponding stages of the short-jointed 

 variety this great change of form has not taken place, thougli the size of the joints lias 

 considerably increased. 



It is difficult to account for these striking variations in the form of the cirri of 

 Antedon phalangium. We have here to deal with something more than a i imor- 



phism of the cirri in any given individual, such as I have noted in the boreal Comatulaa 

 Antedon hystrix, A.prolixa, and^. quadrata, &c. ; for the geographical distribution h 

 be considered as well. While the two intermediate types of cirrus occur in individuals 

 from every locality, northern and southern alike, the long-jointed form characteristic of the 

 Mediterranean is not to be found in any of the Scotch specimens; while those from the 

 Mediterranean, the Seine bank, and even from 220 fathoms in the Atlantic; off the 

 Portuguese coast, have none of the quite short-jointed cirri, such as thai shown in 

 fig. 33. Neither is it in any way a bathymetrical variation. The Scotch specimens 

 were obtained in the Minch at a depth of 60-80 fathoms. The ' Dacia dredged on the 



SECOND SERIES. — ZOOLOGY, VOL. II. 



