290 Beer. Development of tlie pollen graiii and anther of some Onagraceae. 



raetliylono l)luo and fiiclisin aro all descrihcfl as pectic stains 

 whicli leave callosc uiicoloured. 1 liave found thoiii all to stain 

 the iiiothor-coll wolls althou<^h not nearly so deoply as the peetic 

 membranes. 



The origen of callose has formed the subject of rej^eated 

 disc'ussion. In the case of the callose of tlie Sieve-tvibes some 

 have asserted that this substance arises from the transformation 

 of pre-existing cellulose whilst others believe it to be a direct 

 product of protoplasmic activity originating withont any relation 

 to cellulose or other fore - runner. Hill in liis account of the 

 sieve-tubes of Pi)ius-, takes up an intermediate position and 

 believes that callose may originate sometimes directly and 

 sometimes indirectlv ^). 



In the case of the callose composing the pollen mother-cell 

 walls there can be no doubt concerning its mode of origen. 



It has already been mentioned that in the very young 

 anther the walls of the primary sporogenous cells are poorer in 

 cellulose than the other tissues of the anther. In somewhat 

 older anthers, but still long before the mother-cell wall may he 

 expected to appear, the membranes of the sporogenous cells no 

 longer show anj traces of cellulose. Even after treatment with 

 dilute acid and alkali — as recommended byM angin — I was 

 imable to demonstrate any cellulose in these walls. 



It is within these walls that the callose layer is developed. 

 There is, here, no disappearance of either cellulose -j or pectose 

 to account for a transformation of these substances into callose. 

 Whatever may be the explanation of the formation of callose 

 in sieve-tubes. I think there can be no doubt that in the case 

 of the pollen-mother-cells the callose is derived directly from 

 the activity of the protoplast witliout the intermediation of 

 cellulose. 



Each mother-cell now divides to form the four special- 

 mother-cells. The mitotic figure is rather small and not well 

 adapted for studying the details of nuclear division. 



I will content myself, therefore, with stating that in Ocno- 

 tlicra lo)igiflora the number of chromosomes wliich appear at 

 the first and second divisions of the pollen mother-cell is seven. 

 The}^ are so small in size that I can only distinguish them as 

 somewhat irregulär granules; whether they have a definite and 

 constant shape peculiar to each division (as seems likelyi could 

 not be certainly determined (Figs. 5—9). In the somatic divi- 

 sions (which I have studied in the wall-cells of the anther) the 

 chromosomes have the form of curved rods which are crowded 



1) „The Histolog\- of the sieve-tubes of Pinus-'. (Ann. of Bot. Vol. XY. 

 1901. p. 597.) 



■^) The cellulose is lost sight of in the walls of the sporogenous cells 

 far too long before the oallose appears for these substances to have any 

 ciinnection with one another. Moreover the cellulose which occurs in the 

 young sporogenous wall is merelj- a trace and could not possibly account 

 for the massive callose wall. 



