452 SUMMARY OF CURRENT RESEARCHES RELATING TO 



theory that all gamophyllous whorls arc the result of cohesion, this 

 cohesion sometimes taking place in a very rudimentary condition of the 

 organs. It may tako place either in the collateral or in the serial 

 direction, or in a combination of the two. The inferior ovary must be 

 regarded as resulting from the serial coalescence of the members of 

 the different whorls. If it is an axial structure, then all ganiophyllous 

 whorls must be considered to be tubular differentiations of the axis. 

 All placenta) are organs of a foliar character. 



Size and Colour of Alpine Flowers.* — From observations of a 

 largo number of Alpine flowers, Herr E. Keller has come to the conclu- 

 sion that their larger size is generally comparative only to the size of 

 the plant, not absolute ; and that their conspicuousness is due to the in- 

 tensity and other peculiarities of their colouring, such as the scarcity of 

 white and yellow flowers as compared to red. The number of species of 

 insect is not less than at low levels, but the number of individuals is 

 very much smaller, and the favourable time for visiting flowers shorter. 

 The greater intensity of colour of the flowers is partly due to a physical 

 effect of light, which can be demonstrated by experiment. 



Trigger-hairs of the Thistle-flower.^— Prof. B. D. Halsted describes 

 the structure of the hairs found upon the filaments of the stamens of 

 Cnicus altissimiis Willd. Each trichome consists of two nearly parallel 

 cells, which extend side by side to nearly the end of the outgrowth. 

 There is a hyaline outer layer common to the two cells. 



It is not difficult to determine the origin and development of these 

 twin-celled hairs if the filaments are taken for study while young. The 

 surface is at first smooth ; in slightly older stamens small enlargements 

 at certain places, where the surface cells meet end to end, may be recog- 

 nized. The ends of these two cells now take on a lateral growth, and 

 soon become bent at right angles to the surface of the filament. 



These trichomes, therefore, originate by the lateral extension of the 

 ends of two adjoining cells, and they evidently play an important part in 

 the movements of the filaments. 



Ovules of Plantago.J — Prof. H. Baillon corrects previous descriptions 

 of the position of the ovules in this genus. It varies remarkably in the 

 different species. In P. alpina and maritima "one of the two loculi 

 incloses a single ovule at the upper part of the septum ; the other con- 

 tains two at the base. In P. maxima there are usually two ovules in 

 each loculus a little above the base. In P. arabica, Lagopus, saxatilis, 

 Gijnops, aristata, lanceolata, and Webbii, there is one in each loculus above 

 the middle. In P. coronopus there are two in each loculus, or four, of 

 which two are imperfectly developed, or three, of which two stand lower 

 than the others. In P. subulata the young pistil has three ovules in each 

 loculus, one above the others and on the central line, the other two lower 

 and lateral, with their backs turned to one another. In P. major there 

 are 12-15 ovules in each loculus, or sometimes fewer, placed irregularly 

 in several rows. 



* Keller, R., ' Die Bliiten alpiner Pflanzen, ihre Grosse u. Farbenintensitat,' 

 36 pp., Basel, 1887. 



t Bull. Torrey Bot. Club, xv. (1888) pp. 82-4 (2 figs,). 



% Bull. Mens. Soc. Linn. Paris, 1887, p. 663. See Bot. Centralbl., xxxiii. (1888) 

 p. 10. 



