TRANSACTIONS OF SECTION D. 699 



separated from the inner in cross-sections, and the pores were frequently obliterated 

 by numerous very fine granules contained in it. In the ripe egg of the cod ( Gadus 

 morrhua) only one thin membrane is visible, which, as far as I could ascertain, 

 is not porous. I am not quite convinced that the pores are really absent in this 

 case, as I only examined spirit specimens, but it occurred to me that what we see 

 here is really only the outer denser part of the zona radiata, the inner portion 

 having become absorbed during development. That such a thing might happen 

 has been fully demonstrated by Balfour in the e^^ of Scyllium. Balfour's vitelUue 

 membrane is equivalent to the outer part of the zona just mentioned. Both of 

 these seem to become absorbed in some of the Elasmobranch eggs during later 

 development. 



With regard to the Teleosteans I might mention that, in case the presence of a 

 true vitelline membrane should be definitely established in all intra-ovarian eggs, 

 the zona radiata is to be regarded as a cuticxdar formation of the ovum. 



Every intra-ovarian egg is surrounded by the follicle or granulosa, which is a 

 cellular layer. In a paper which I propose to publish shortl}^, the formation of 

 the yolk will be I'uUy considered along with some remarkable changes which take 

 place in the nucleus of the growing egg. 



TUESDAY, SEPTEMBER 7. 



The following Papers were read : — 



1. On Humboldtia laurifolia as a Myrinehopliilous Plant. 

 By Professor F. O. Bower. 



It is already well known that the hollowed and swollen internodes of Hum- 

 boldtia laurifolia are inhabited by smaD black ants. The questions which present 

 themselves with regard to this symbiosis are — 1st, How do the hollows originate ? 

 2nd, Is the presence of the ants of any advantage to the plant ? An investigation 

 of young shoots shows that the opening, through which the ants enter, is formed 

 by rupture of the superficial tissues, owing apparently to pressure from within, 

 and that the ants thus gain access to and hollow out the pith which had previously 

 begun to decay. Thus the plants take the initiative, and the ants are not slow to 

 avail themselves of the opportunitj^ oti'ered. Further, from the numerous glands 

 on the leaves it is probable that they derive food, and are thus supplied with both 

 nourishment and lodging. No evidence is forthcoming, however, that the symbiosis 

 is of any advantage to the plant. The stipules in this plant are of a peculiar form ; 

 a study of their development shows that a peculiar auricle-like outgrowth is formed 

 at the base of the simple, young stipule, subsequently to the origin of the latter. 

 Though this assumes a peculiar, almost sagittate form, still in its real nature it is 

 similar to those auricles which are formed at the base of the stipules of Fiola 

 tricolor. 



2. On Positively Geotropic Shoots in Cordyline australis. 

 By Professor F. O. Bower. 



It was noticed in Peradeniya Gardens that when, by reason of the weight of 

 the head of leaves, stems of Cordyline australis assumed an oblique or horizontal 

 position, shoots were formed from the lower side pointing directly downwards. It 

 was ascertained that these were in their origin axillary. By hanging pots of soil 

 in such a position as to immerse the tips of these shoots, an elaborate root system 

 was soon formed, the roots arising in the usual manner. In the cases observed the 

 apex of the shoots remained covered with scale leaves, and the ultimate fate of it 

 is uncertain. It is clear that here we have a special adaptation for mechanical 

 and physiological support of a weakly axis, and in this respect we may compare 



