OF SANTALUM ALBUM. 75 



vertical position (fig. 33), or with the cotyledons curved a little towards the lateral 

 basUar scar just mentioned (fig. 35). The radicle is at the upper end of the seed and 

 terminates in an acute apiculus (Tab. XVIII. figs. 34 & 36). 



In the mature fruit the mesocarp forms a hard shell, outside which the epicarp forms a 

 thin layer of pulpy substance (Tab. XVIII. fig. 22). The woody shell is slightly pointed 

 and trigonal above, presenting three converging ridges (Tab. XVIII. fig. 31) when the 

 epicarp is removed. Witliin the woody mesocarp the albvimen or endosperm of the 

 seed lies free ; the coat formed by the embryo-sac is no longer distinguishable, but the 

 endosperm is covered with brownish membranous scales (Tab. XVIII. fig. 32), composed 

 of compressed withered fragments of the lax cellular tissue of the obliterated parenchy- 

 matous endocarp. 



The above observations confirm in almost every respect those published by Griffith in the 

 Transactions of the Linnean Society, the main point of difference lying in the statements 

 made with respect to the phtenomena presented at the summit of the embryo-sac at the 

 time of fertilization ; in which my account is strongly opposed to those given by that 

 author, not only in Santalum, but in Osyris and the Loranthacece. Notwithstanding 

 the high value I attribute to Griffith's labours, increased and confii-med by the researches 

 now brought forward, I feel very confident of the correctness of the account I have given 

 of the origin of the embryo from a pre-existing germ, and I have little doubt that the 

 process of fecundation is such as I have described in Osyris, and the other cases. The 

 importance attaching to the truth of the view I have propounded will be farther illustrated 

 below. 



Every one who has studied the development of the ovules of Santalum, and the allied 

 genera, has been struck by the remarkable anomalies which present themselves. The 

 entu'e protrusion of all the (apparently) essential part of the embryo-sac from the apex 

 of the nucleus, the development of the endosperm in the external compartment of the 

 sac, altogether independently of the nucleus, are very remarkable ; while the posterior 

 development of the embryo-sac is nO less singular. The idea has been suggested that the 

 entii-e central body here described as a free placenta with the ovules reduced to nuclei, 

 might be one ovule with three embryo-sacs ; and also that the central body is all placenta, 

 the nuclei being merely the funiculi of ovules reduced to embryo-sacs. There does not 

 seem to be sufficient grovmd for either of these assumptions, although the former is in 

 some respects plausible. 



The principal reason which could be advanced in favour of the second idea, is the appa- 

 rently abnormal position of the embryo, the cotyledonary extremity of which is next the 

 apex of the nucleus, and the radicle end pointing in the direction of the base of the 

 nucleus. But this is rather an apparent than a real ii-regularity. It would be a pedantic 

 insistence upon terms to call that end of the embryo-sac engaged in the apex of the 

 nucleus, the micropyle-end ; it is really the middle, and the embryo-sac is in fact campy- 

 lotropous, its organic base or chalazal end being in the interior of the nucleus, while the 

 micropyle-end is prolonged out beyond the micropyle, and turned up so as to Ue (out- 

 side) against the chalazal end. 



In regard to the fii'st of the two views above referred to, the only cii'cumstances which 



l2 



