Osyris, Loranthus and Viscum. 179 



scribed as "ovarium" (Tab. XIX. fig. 1); in the other, exemplified by Lo- 

 ranthus bicolor of the same author, they extend a long way up the canal of 

 the style, reaching not unfrequently to within a short distance from the stig- 

 matic surface itself. 



In both modifications, after obvious action of the pollen on the stigma, 

 filaments will be found adhering to the heads of the sacs, into which they 

 penetrate, becoming dilated within the dilated apex of the sac, and constricted 

 in its constricted part ; not, however, to their limits or diameters previous to 

 their penetration (L. bicolor, Tab, XX. fig. 6 ; L. globosus, Tab. XIX. fig. 2) : 

 they run down throughout the whole length of the sac, being always, so far 

 as I have seen, while within this organ, two in number and in close appo- 

 sition to each other. In L. bicolor they appear to run without interruption 

 of continuity (Tab. XXI. fig. 2) almost to the very end of the sac ; in L. glo- 

 bosus they have generally appeared jointed immediately above that part of 

 the sac in which the albumen becomes developed. I cannot state posi- 

 tively whether these tubes, or their similar continuations, pass out through 

 the lower end of the sac, but growths from the ends of these, shorter, of 

 larger diameter, and subsequently of increased number, certainly appear to 

 do so. 



In that modification of sac which does not extend beyond the so-called 

 ovarium, the penetration of the tubes appears almost immediately associated 

 with the appearance of grumous nuclei in a particular part of the sac, which 

 nuclei, soon becoming invested with proper membranes, form the cellular 

 tissue of the young albumen. But in the other modification this is not the 

 case, but the analogous cellular tissue is developedyrow the ends of the sacs, 

 and within the mass there formed the continuations of the pollen tubes will be 

 found. (Tab. XXI. fig. 3.) And it is remarkable, that whereas in the first 

 modification the sac retains its original simple membranous nature, unless 

 pollen tubes have penetrated into it {L. globosus. Tab. XIX. fig. 4), yet in the 

 second, each sac appears to have to some extent independent powers of growth, 

 inasmuch as cellular tissue is often produced from their ends without any 

 such penetration having taken place. (L. bicolor, Tab. XX. fig. 7 ; Tab. XXI. 

 fig.l.) 



In Loranthus bicolor the continuations of, or growths from, the pollen 



VOL. XIX. 2 b 



