QUEKETT MICROSCOPICAL CLUB. 353 



modes of division. Simultaneous division where the two nuclear 

 divisions occur before any walls are found is more characteristic 

 of dicotyledons. In the successive method, most frequent amongst 

 monocotyledons, a wall follows the first nuclear division, forming 

 two hemispherical cells, which again divide equally. A very 

 variable amount of the hypodermal layer of the young anther 

 may become archesporium. The primary sporogenous cells 

 directly or by division produce the mother cells ; in some cases 

 they become mother-cells without division, and in some orchids 

 each primary sporogenous cell forms a well-defined mass of mother- 

 ceUs, a massula separated from its fellows by thicker walls. The 

 ultimate form of the pollen grains is to some extent determined by 

 these differences in the method of formation. Irregularity in the 

 number of divisions of the mother- cell may give rise to more or less 

 than four microspores. Dr. Eendle described several ways in 

 which this may take place. Each of the four young microspores 

 (pollen grains) becomes invested by a separate wall. This soon 

 becomes differentiated into two layers, the inner of pure cellulose, 

 and later developing the pollen tube, the outer cutinised and often 

 sculptured, generally leaving spots for the exit of pollen tubes. A 

 single point of exit occurs in most monocotyledons and a few 

 dicotyledons, while in most dicotyledons there are two or more 

 points of exit. In the family Cucurbitaceae and in the Passion 

 flower lid-like pieces of the outer wall become detached. The 

 origin and development of the walls of spores is a problem that 

 needs further investigation. The character of the pollen is in 

 some cases determined by the ultimate degree of separation be- 

 tween the grains. Generally they are entirely free at maturity, 

 forming a powdery mass, but in some cases they cling together, 

 and in some orchids and the asclepiads they form one mass, the 

 poUinium. In water plants, when pollination takes place 

 beneath the surface, the grains are thread-like, while in wind- 

 fertilised plants they are small and dry, the sculptured and 

 adhesive grains being characteristic of insect-pollinated plants. 

 Thus there is some relation between the form of the pollen 

 grain and the habit of the plant, but when the influences 

 of method of development and habit are eliminated there 

 still remains a wide field for investigation as to the meaning or 

 causes of the wonderful diversity of the mature pollen grain. In 

 the case of early flowering herbs and catkin-bearing trees the 



