444 DR. S. F. HARMER ON THE 



obtainerl towards the end of October, and therefore soon after the 

 hibernacula had been formed. It is not improbable that they 

 wonld ail liave become darlver as tlie season progressed. All were 

 growing on erect branches of the colony, and most of them were 

 quite free, although some show the habit which has been described 

 by other observers of attaching themselves to foreign objects ; 

 in this case zocecia, or even other hibernacula, of the Paludicella 

 itself, or zocecia of the Fredericella with which the species was 

 associated. Allusion has already been made to the fact that one, 

 two, or even thi'ee hibernacula are boi-ne on nearly every zooecium. 

 They vary much in length and to a less extent in breadth. 

 Leaving out of account the two hibernacula of which one is 

 shown in fig. 1, they are nearly all spindle-shaped, although 

 occasionally showing some irregularity of form, and vary in 

 length from 608 ju (fig. 4) to about 1968//. They are usually 

 pointed at their free end, the broadest part being as a rule beyond 

 the middle of their length. The proximal stalk-like portion is 

 generally slender, but vai'ies in breadth from about 72 p. to 96 p. 

 Measured at their widest part, the width varies from 144/,(to 

 160/x. They are usually developed in the position occupied by 

 either the lateral or the terminal daughter-zooecia ; but, as shown 

 in fig. 7, the hibernaculum may originate more proximally than 

 the position normally occupied by a lateral bud ; and the bases 

 of the lateral hibernacula ai'e not always opposite one another 

 (fig. 8). Most of the hibernacula contain an elongated polypide- 

 bud, in agreement with Kraepelin's statement (92), of the form 

 which also occurs in the development of the polypide-buds of 

 ordinary zocecia, I have not observed any calcareous matter in 

 their wall. 



It has often been suggested that the statoblasts of Phylolacto- 

 ]8ema,ta may have been evolved from structures like the hiber- 

 nacula of Paludicella {cf. Kraepelin, 87, pp. 163, 167; 92, p. 62). 

 But while the mode of origin of the statoblasts, from the funi- 

 culus, is well understood in its general outlines, I have not been 

 able to discover any account of the development of the hiber- 

 nacula. The material at present under discussion enables me to 

 make a contribution to this subject, although the facts could have 

 been interpreted with more certainty if the specimens had been 

 collected slightly earlier in the year. There is, however, enough 

 evidence to show that the hibernaculum is not a modified terminal 

 zocecium, as is assumed by Braem (90, p. 112) in criticizing 

 Kraepelin's opinion, but is something which is left when the 

 distal part of a zocecium has been thrown off'. The evidence of 

 figs. 2, 3, '9, 10 seems to be unmistakable on this point. In these 

 cases the hibernaculum is well formed, with its thick cuticle and 

 inclusions of yolk-like material. But although its stalk appears 

 to be constituted, in some cases at least, by the proximal part of 

 the zooecium, the distal end of the hibernaculum lies freely in the 

 cavity of the zooecium, which is prolonged in its normal fo>m 

 beyond it. The interpretation which natui-ally suggests itself is 



