30 CELL-FORMATION. [BOOK i. 



Molil. From the observations of this anatomist it appears 

 that in Confervse the increase in number of the cells takes 

 place by the internal division of a parent cell. In Conferva 

 glomerata the last joint is always as long as those below it, 

 only rather more slender. The branches grow at the upper 

 lateral extremity of a joint or cell ; each at first is a small 

 protuberance, which is transformed into a lateral cylindrical 

 excrescence containing chlorophyll (green colouring matter), 

 and having its cavity in communication with that of the joint 

 which bears it; as the branch lengthens, a contraction is 

 observable at the line of insertion, which contraction is 

 directed towards the interior of the cell, and chokes up the 

 green matter, forming a sort of partition, pierced in the 

 middle like a ring. This partition grows with the growth of 

 the branch, and at last completely cuts off all communication 

 between the first cell and its branch. Thus cut off, the latter 

 lengthens by degrees, till it forms a very long cylindrical cell, 

 which divides in just the same manner into two other cells, the 

 terminal of which alone lengthens, to be bisected in its turn. 

 Mr. Henfrey, in an able paper read before the British 

 Association, Sept. 1846, produces new evidence in support 

 of this opinion, viz. : " That the division of a parent cell 

 into new cells is effected by the gradual folding inward of 

 the primordial utricle, which organ, in virtue of its peculiar 

 function, secretes the septum within that fold ; the circular 

 constriction thus produced arriving finally at the centre, the 

 septum consisting of a double layer of cell-membrane becomes 

 complete." The evidence upon which he relies is stated 

 thus : " In the course of my investigations to satisfy myself 

 of the correctness of the view I had taken of the agency 

 exercised by the primordial utricle in cell-division, I have 

 observed the process in several plants, Cryptogamous and 

 Phanerogamous. In no case have I been able to trace the 

 gradual progress of the formation of septa so well as in 

 Achimenes grandiflora. This plant produces a great number 

 of axillary buds or bulbets, on the scales of which are found 

 many capitate hairs. I examined these hairs in young buds 

 of from about half a line to a line in length, possessing at 

 that period only six or seven scales. By dissection these 



