Dr. W. Hofmeistcr on the Fecundation of the Conifers?. 433 



least difficulty in observing these cells ; in the Abietinese and 

 the Cupressinese, on the contrary, their outlines are often diffi- 

 cult to detect, on account of the fluid contents of the corpusculnm 

 and its daughter- cells refracting light in exactly the same de- 

 gree. But by keeping the preparation for a longer time in 

 water, still better by treating it with a very dilute acid, all the 

 conditions come out clearly even in these. 



The internal walls of the upper part of the corpuscula of the 

 Cupressinese become thickened, often very strongly, in the man- 

 ner of scalariform vessels. In Biotia orientalis, in particular, 

 these walls when viewed in face present a very elegant appear- 

 ance, through the transverse streaking, depending on the small 

 breadth and close apposition of the slit-like thinner parts of the 

 membrane. 



The formation of the free cells in the interior of the pollen- 

 tube does not commence until its arrival at the albumen-mass, 

 in the Cupressinese (Ju?iipe?'us communis, Sabina and virginiana ; 

 Biotia orientalis), after the rupture and passage through the mem- 

 brane of the embryo-sac (dissolved at the point of contact), and after 

 the hollow above the apex of the corpusculum has been filled up. 

 The earliest conditions (recognized as early by the corpuscula, 

 with untouched rosette lids, still containing a large central 

 vacuole, the formation of free cells in their interior not having 

 yet taken place) display, swimming in the centre of the expanded 

 portion of the pollen-tube, one large, free, globular cell with a 

 very soft membrane, through, the fluid contents of which, ren- 

 dered opake by numerous granules, a central globular nucleus 

 shines as a lighter space. Preparations in which the pollen- 

 tube had commenced the displacement of the rosette of cells 

 covering the corpuscula, frequently exhibited two such cells, or 

 the cell had a longish ellipsoidal form and contained two nuclei, 

 one in each focus. In a few cases I saw three, in two cases 

 four such globular cells in the end of the pollen-tube. When 

 there are more than one, they are always smaller than a solitary 

 one. Other pollen-tubes, taken from ovules where the corpus- 

 cula were filled with daughter-cells, contained in place of one or 

 more of these large cells, one or two groups of four to eight 

 smaller cells, with similar contents to the larger, adherent to the 

 wall. Sometimes a number of such smaller cells occur with one 

 large one in the same pollen-tube. In many cases the smaller 

 separate from one another when the preparation is pulled about 

 in extracting the pollen-tube. When these smaller cells are 

 burst by pressure, three things may be distinguished in the con- 

 tents : very small granules, globular or angular, in which the 

 three dimensions are about equal ; longish corpuscles, spindle- 

 shaped or stick-shaped, coloured brown by iodine, like the 



Ann.^ Mag. N. Hist. Ser. 2. Vol. xiv. 28 



