644 Journal of Agricultural Research voi. xii, no. 10 



The; tapetum. — The tapetum, usually one cell layer thick, is com- 

 posed of large angular cells similar in size, and staining reactions to the 

 pollen mother cells. The tapetum in Fragaria spp. differs in its greater 

 persistence from most other forms reported. Division in the tapetal 

 cells begins at about the time of synapsis of the pollen mother cells and 

 has been observed as late as the metaphase of reduction division. The 

 divisions are all mitotic, no evidences of amitotic divisions having been 

 observed. Following reduction division the tapetal cells are binucleate 

 and remain so until the liberation of the microspores, when they degen- 

 erate and completely disappear. The disappearance of the tapetal 

 layers is gradual. They first separate from the wall layers and then 

 proceed to dissolve, the wall which was in contact with the anther wall 

 first disappearing, followed by gradual dissolution of the entire cell 

 layer. During this process the walls become thick and laminate and in 

 places the middle lamella dissolves, partially freeing the individual cells. 

 When the microspores are three-quarters grown the tapetal cells have 

 entirely disappeared. This persistence of the tapetal cells will be shown 

 to be correlated with a like persistence of the pollen mother-cell walls. 

 The history of the tapetum in F. virginiana is identical with that just 

 described which is of Minnesota 3. 



Pollen mother cells. — The study of pollen development was begun 

 with the so-called resting stage of the pollen mother cells between the 

 last archesporial division and the first meiotic division. They do not, 

 however, show the characteristic chromatin and linin condition found 

 in true resting cells of Fragaria spp. (compare fig. i, PI. B and fig. 2, 

 PI. D). The cells are angular and contain a large nucleus (PI. B, i). 

 The contents of the nucleus are irregular dark-staining, very small masses 

 of chromatin held in a network of linin fibers. The number of chro- 

 matin bodies is very much larger than the number of chromosomes. 

 Ordinarily one large nucleolus is present, although it is not uncommon 

 to find two. 



Synapsis. — ^The first indication of the onset of the prophase is to be 

 seen in the gradually increasing size of these chromatin bodies which 

 still, however, appear very flaky and irregular. The linin and chromatin 

 during this period are so indistinct that it is impossible to determine 

 whether there is any definite pairing of the individual particles and 

 threads as has been shown to exist in Lilium spp. by Allen (2) and in 

 several forms by Overton (jo). These larger masses gradually move to 

 one side of the nuclear cavity and congregate about or near the nucleolus 

 in a loose indefinite mass (Pi. B, 2). At this time a few rather defi- 

 nite threads appear in the mass, some extending out from it as loops. 

 Where the loops are long enough, they are seen to be distinctly double 

 (PI. B, 2, 3). This is the only evidence of any pairing during the 

 presynaptic stages. Gradations between the conditions shown in 

 figures I and 2 occur within a single loculus of an anther and are proof 



