484 SUMMARY OF CURRENT RESEARCHES RELATING TO 



ones closely united, or two of nearly equal size. It is ofteu evident that 

 the spirem thread is connected with these nuclear parts. On the dis- 

 appearance of the nuclear memhrane in Symplocarpus the kinoplasmic 

 fibres entering the hollow are quickly oriented as a multipolar sj)indle 

 with rather indefinite apices. The next stage is a multipolar spindle with 

 very few poles, followed finally by a bipolar spindle. In Peltandra 

 there is a less definite multipolar spindle than in Symplocarpus, although 

 the general form is evident. On the dissolution of the membrane of the 

 microspore mother-cells, the tapetal cells become free, and wander in 

 between the maturing microspores ; they lose their identity, and become 

 a protoplasmic stratum in which the microspores are imbedded. In 

 Symplocarpus there is no division of the generative nucleus in the 

 pollen-grain. The vegetative nucleus usually passes first into the 

 pollen-tube. The generative nucleus passes into the tube while still 

 enclosed within its cell-membrane. 



Cross-Pollination and Self-Pollination. — The late Mr. G. W. Ord * 

 gives some interesting particulars respecting the visits of lepidoptera to 

 flowers. With regard to moths, he believed that colour is of little im- 

 portance in attracting them ; they will find their way even to inconspicu- 

 ous flowers by the sense of smell; white flowers seem to be a hindrance 

 rather than a help even to night-flying moths. 



All the species of Bomulea are, according to Sig. A. Beguinot,f pro- 

 terandroup, and are adapted for entomophily, but can be self-pollinated 

 when this fails. The flowers are remarkably polymorphic, varying in 

 colour, in the size of the perianth, in heterostyly, and in polygamy. 



Prof. G. Arcangeli J describes the facilities, for the carriage of pollen 

 in several species of Cucurbitaceaa, with the insects which assist in the 

 process. In Luffa cylindrica, which has extra-floral as well as floral 

 nectaries, the same kind of insect was in no case observed to visit both 

 kinds of nectary. 



Mr. J. G. Needham § gives an elaborate account of the floral biology 

 of Iris versicolor, the contrivances for facilitating the obligatory cross- 

 pollination by insects, the insect-visitors, and the mode in which hurtful 

 insects are misled by the deceptive colour-streaks on the petals. He 

 believes the efficient insects, mainly bees, to be largely guided in their 

 visits to the nectaries by the coloration of the flowers. 



Law of "Splitting" of Hybrids.!) — M. H. de Vries lays down the 

 law that a hybrid never " halves " any of the special characters either 

 of the father or of the mother. It inherits some characters of the one, 

 some of the other parent. When the parents differ on only one point, 

 the hybrid does not occupy an intermediate position, but resembles one 

 or the other parent. In the hybrid, the simple differential character of 

 one of the parents is evident or "dominant," while the antagonistic 

 character of the other parent is latent or " retrogressive." These antago- 

 nistic characters usually remain combined during the whole of the 

 vegetative life of the hybrid, the one dominant, the other latent ; but in 

 the reproductive period they become separated ; each pollen-grain and 



* Trans. Nat. Hist. Soc. Glasgow, v. (1900) pp. 355-66. 



t Bull. Soc. Bot. Ital., 1899, pp. 214-22. J Tom. cit., pp. 198-204. 



§ Amer. Natur., xxxiv. (1900) pp. 361-86 (1 pi. and 4 figs.). 



j| Coniptes Rendus, exxx. (1900) pp. 845-7. 



