196 Mr. Griffith on the Ovulum of Santalum, 



than such an ovulum as he describes may exist, as appears to me to be pointed 

 out by tantalum and Thesium. The nucleus of an ovulum is the part first 

 formed : in the direction of its growth it obeys those laws that regulate all 

 extensions of the axis, its apex being formed first, and once formed, always 

 continuing to be the apex, and it is always as compound in its commence- 

 ment as any direct extension of the axis is at its commencement. None of 

 these conditions are fulfilled by the sacs in question ; and in Viscum album 

 the direction of growth appears to be quite reversed. 



There are also two other circumstances, which, I think, are corroborative of 

 my opinion, namely, the development of the albumen in the interior of the 

 sacs of Viscum and L. globosus, and the absence of a sac surrounding the 

 embryo ; circumstances which, taken together, are, so far as I know, contrary 

 to every analogy. 



Considering, then, their late development, their structure, which is that of 

 almost all albuminigerous embryo-sacs, and which is so unlike that of a nucleus, 

 their similarity in the same respect to the embryonary sac of Santalum, The- 

 sium, Osyris*, &c, I have little doubt but that they are the^analogues of 

 ordinary embryo-sacs, although I am by no means certain, from not having 

 seized on their first development, that the anomalies may not be so great as I 

 have conjectured: for to each sac there may be a nucleary base, or indeed a 

 common one; for I see no reason why exsertion should not occur in nuclei 

 with several embryo-sacs as in those with only one. 



With the exception of the structure of the sacs, and their perforation lon- 

 gitudinally by the pollen tubes, the sacs of the two species have not much in 

 common ; for while in Loranthus globosus they have not very extraordinary 

 limite, ,n L^bicolor they are found a long way up the stigmatic canal ; and 

 wh.le m the former species the albumen has its usual relations only partially 

 interrupted, in L. bicolor these appear almost, if not quite, entirely destroyed ; 

 and what is very remarkable, the sacs have to a certain extent proper powers 

 of productive growth. Cellular subdivisions likewise make their appearance 

 in both species above the albumen , and consequently in L. bicolor, in the sac 



itselt. I am unable to state whether these really an * a • *■* *•. 



c wueuiei mese lealty enter into the composition of 



* The sacs of Viscum album appear to disagree remarkablv Jr, ™* k ■ 

 simple cell. remarkably m not bemg Simple extensions of a 



