124 DR A. MILNES MARSHALL ON THE 



Koren and Danielssen, Whiteaves and Verrill, maintain the specific 

 distinctness of this form from P. phosplwrea. The measurements of the 

 different specimens I have been able to examine show so great variability in 

 the essential characters of the form, such as the length of the large zooids, 

 width of leaves, &c, that I can have no hesitation in agreeing with Kolliker* 

 in regarding it as merely a variety, though a very well-marked one. I cannot, 

 however, accept RicmARDi'st conclusion that it is not even a variety, " ma uno 

 stato puramente accidentale di certi exemplari." 



The general appearance of the form is shown in PI. XXI. fig. 4, representing 

 an entire specimen seen from the right side and twice the natural size. Fig. 5 is 

 a transverse section through the rachis about the middle of its length, with one 

 of the attached leaves and the base of the corresponding leaf of the opposite 

 side. For the sake of comparison, I have given in fig. 6 a similar view of a 

 normal specimen of P. phosplwrea obtained from Oban. 



If these two figures be compared together, it will be seen that the points in 

 which the variety aculeata (fig. 5) differs from the typical P. phosplwrea (fig. 6) 

 are the following : — The leaf is longer and much narrower ; the polyps are fewer 

 in number, are placed further apart, and are independent of one another for a 

 greater portion of their length. The walls of the rachis are much thicker, a 

 condition associated with the presence of larger and more numerous spicules ; 

 the axial calcareous stem is thicker, and the main longitudinal canals of the 

 rachis much smaller than in the typical form. Concerning the zooids, it will be 

 seen that a very large one (fig. 6,/) arises from the rachis immediately below 

 each leaf, with the ventral border of which it is fused for about a third of its 

 length. Nearer the mid-ventral line of the rachis there is on each side a second 

 row of large zooids, usually slightly smaller than those of the outer row, but 

 sometimes equalling them in size; and within these again other zooids occur 

 intermediate in size between the large ones and the normal small ones. The 

 largest zooids, those attached to the ventral borders of the leaves, have an 

 average length of rather over 3 mm. in the "Triton" specimens; in exceptional 

 cases they reach 4 mm. 



Between the large zooids are numerous small ones of the normal size and 

 character, which extend up the sides of the rachis between the bases of the 

 leaves. 



That the connection of the large ventral zooids with the leaves is of a purely 

 secondary nature is clearly seen by tracing their gradual development in passing 

 upwards from the lower end of the rachis. At their first appearance they 

 are small, identical in all respects with the normal zooids, and quite inde- 



* Kolliker, Op. cit., pp. 134 and 366. 



f lticuiARDi, Monoyrafia delict famiylia del Pennatularii, p. 24. 



