783 



opening, and found that the lower portion of the tube on the former 

 (the full-blown flower) had generally became tubular to the amount 

 of from eight to ten cells in each hair, which usually consisted of from 

 twenty-five to thirty cells ; whilst that of the latter (the opening flower) 

 had become tubular only to the extent of about two or three cells. It 

 therefore seems that the time occupied between the first opening of 

 the flower and its fading period is suflicient to convert six or eight 

 cells into tube ; and it is probable that in the earlier part of the sea- 

 son, during more genial weather, the fading stage would be delayed, 

 and consequently that in proportion as the blooming period is pro- 

 longed, the length of the tubular formation would be increased ; but 

 it seems doubtful whether these tubes ever become tubular through- 

 out their entire length. 



" On the two contiguous petals, on the inner and inferior part of 

 each, on a prominence where the claw takes its origin, there is a ridge 

 of hair of a totally different character from that on the claw of the 

 spurred petal, being at all ages pervious throughout, dilating gradu- 

 ally from its base to within a trifle of its extremity, when it again gra- 

 dually lessens in diameter, until it terminates almost spherically. 

 There is not any second cell to be detected in any portion of these 

 tubes, even before the flower opens. Their origin and termination 

 seem to be a simple cell, lengthening and dilating ; and therefore they 

 are clearly unicellular. In the multicellular tube it is evident that, in 

 order to secure a tubular structure, Nature clearly manifests her 

 intention, by generating a single linear series of cells ; and that this 

 multicellular tube shall be a cone she as clearly manifests her determi- 

 nation, by generaring cells gradually decreasing in transverse diameter 

 from the base to its apex ; and it would seem that where she has com- 

 pleted her cellular arrangement as regards their position and forma- 

 tion, her subsequent care is, by some peculiar and amalgamating 

 process, to unite the adjoining cell-walls into one compact septum, 

 denominated a partition wall. 



" As regards this septum, it is not unreasonable to hope that 

 repeated and minute microscopical investigations of the progressive 

 growth and formation of the multicellular tube, at different periods of 

 its age, may furnish material data on which to found a knowledge of 

 the probable mode of its absorption, or removal. 



" To discover whether the comparative increase of growth of the 

 opening and fading flower kept pace with cellular conversion into 

 tube, 1 measured the transverse diameter of the tubular portions of 

 the two stages of growth of a cultivated plant, and found that the ave- 



