INVERTEBRATA, CRYPTOGAMIA, MICROSCOPY, ETC, 303 



Many of these bear on the upper side a conical outgrowth, obviously 

 the metamorphosed nucleus, though this is as frequently absent. 

 Between this extreme and the normal condition are all sorts of inter- 

 mediate stages. A careful consideration of the phenomena presented 

 convinced the author that the two integuments together of the ovule 

 represent the lamina of the leaflet. The ovule is, in fact, transformed 

 from a simple pinna of the carpel, which produces out of itself a 

 " metablast " in the form of an emergence or trichome. In one or 

 two remarkable instances the leaflets bore on their dorsal side two or 

 more inner integuments instead of one. The specialty of Cela- 

 kovsky's interpretation of the leaf-like ovules of Hesperis is that he 

 considers each leaflet as consisting of two parts — an upper and a 

 lower — representing the two integuments. With respect to the ana- 

 logy with vascular cryptogams, he regards the ovule with its integu- 

 ments as homologous with the pinnule, sorus, and indusium of the 

 Polypodiacete. 



Cell-division.* — Professor Strasburger recommends the hair on 

 the filaments of Tradescantia virginica, or still better of T. elata, as 

 peculiarly favourable for observing the phenomena of cell-division. 

 The nucleus remains clearly visible during the whole course of its 

 division, and all stages of its dijfferentiation can be well seen without 

 the application of any chemical reagent. If immersed in a 1 per cent, 

 solution of cane-sugar in distilled water, the hairs retain their living 

 condition for a considerable time, even as much as twelve hours. 



The nucleus of cells still in a condition capable of division has a 

 diameter of about 0*018 mm. The division takes place especially in 

 the terminal cell of the hair, not unfrequently in the next, compara- 

 tively rarely in those lower down. The protoplasm of the cells 

 contains only very fine granules, and there is therefore nothing to 

 interfere with the observation. 



The nucleus is sharply defined in its whole mass, and presents the 

 appearance of a reticulate structure. While thus sharply diflcrcntiated 

 on the outside, the nucleus has no skin or envelope which is distinctly 

 differentiated towards the interior. It is only rarely that large gra- 

 nules are visible in the living nucleus resembling the nucleoli in 

 form. 



When the nucleus is about to divide, it first of all begins to gi'ow, 

 the diameter in tlie direction of the longitudinal axis of the cell often 

 increasing twofold in length. As soon as the nucleus has reached 

 its full length, changes in its contents are apparent, and protoplasm 

 begins to collect at both its poles. The nucleus now begins to become 

 coarsely granular, and the granules to collect into lines which, of 

 various lengths, permeate the nucleus more or less obliquely, and 

 with more or less of an S-like curvature. The sharp external dif- 

 ferentiation of the nucleus is then lost, and the starch-grains in its 

 interior absorbed. 



From three to four hours have now elapsed since the commence- 

 ment of growth, and the following stages, till the complete formation 



* 'SB. Med.-Naturwisa. (ieaell. Juua,' 1879, p. 'J3. 



