686 NATURAL SCIENCE. ^^v, 



canadensis, while as regards the fall of leaves two methods are to be 

 distinguished. The first, as in Gymnocladus, with a previous cicatrisa- 

 tion of the wound ; the second as in the leaflets of Gymnocladus and 

 leaflets in general, that is with no trace of previous cicatrisation, the 

 wound healing only slowly, remaining for some time exposed as a nidus 

 for spores of parasitic fungi. There are also intermediate cases. 



Further researches are needed to show whether the absorption of 

 the median layer is universal, or whether the separation may take 

 place according to both methods. Strasburger (20), describing the 

 leaf-fall in the horse-chestnut in his " Botanische Practicum " (1884), 

 says, p. 241: "The separation takes place inside the separating 

 layer, the cells of which round off against each other and become dis- 

 connected " ; and Hillhouse, in an editorial note in the English 

 translation of the same, mentions that researches undertaken in 1882 

 at Bonn, under Strasburger's guidance, and still continued, have led 

 to practically the same result. Bretfeld (18) who investigated the 

 process of leaf-fall in the tree-like monocotyledons, orchids, and 

 aroids, finds that the separating layer in these plants is a still earlier 

 formation, being produced at the same time as the leaf itself in the 

 general tissue-differentiation. 



In the case of evergreens, where the leaves are so protected as to 

 be able to withstand the winter cold, the plant is always in leaf. 

 Every season, however, some of the older leaves fall, while new ones 

 are regularly produced. The number of seasons for which they last 

 varies with the species in question ; in Skimmia japonica, a shrub 

 frequently seen in London parks, leaves will be found on the growth 

 produced the year before last, and Mohl says he saw leaves on 

 eleven-year old shoots of the Silver Fir in the Black Forest. 



It is interesting to note that, where the function of the leaves is 

 for some reason performed by another organ, phenomena analogous to 

 leaf- fall may also supervene. In some species of Rubus, the black- 

 berry genus, the leaves are reduced to thorns, and the function of 

 assimilating carbonic acid from the atmosphere is performed by the 

 green cortex of the stem. F. W. Oliver (21) has shown that in Rubus 

 aiistralis the assimilatory layer is cut off annually by a layer of cork, 

 and that a similar case occurs in Casuavina, where the longitudinal 

 ridges on the stem form the chief assimilatory tissue ; these ridges are 

 cut off every year by a cork layer and begin to scale off in the second 

 year. 



It is well known that a sudden frost in autumn often causes a 

 copious leaf-fall. Mohl (15), in mvestigating this point, found a thin 

 sheet of ice on the surface of the leaf-scar, while other leaves which 

 were still attached to the twig were, however, quite separated from the 

 leaf-scar by a similar ice-sheet, to the upper surface of which they 

 were frozen. Closer examination showed the ice-sheet to be formed 

 in the separating layer, the cells of which, being actively living, are 

 more delicate and contain more water than the surrounding tissue, 



