Physiologie. 117 



Kirkwood, J. E., The pollen-tube in someofthe Cucurbt- 

 taceae. (Bull. Torrey Bot. Club. XXXIII. p. 327—342, pls. 16 and 

 17. June 1906). 

 The time elapsing between pollination and the arrival of the 

 pollen-tube at the embryo-sac is twenty-six hours in Melothria , nineteen 

 hours in Micrampelis, and forty-one hours in Cyclanthera. Starch- 

 bearing conducting tissue lines the stylar canal and Covers the 

 placental lobes, and along this tissue the pollen-tube travels, rarely 

 injuring the cells. The cells of the conducting tissue are compared 

 to those of certain nectaries, and like these appear to produce a 

 nutritive secretion which in the present case exerts a directive force 

 on the poUen-tubes. It is concluded that the behaviour of the pollen- 

 tube is a phenomenon of chemotropism. M. A. Chrysler. 



Wopsdell, W. C, The Structure and Origin oi l\).eCycadaceae. 

 (Annais of Botany. Vol. XX. 1906. p. 129-159. With 17 textfigures). 



The paper is divided into two parts. 



I. The habit and structure of the Cycads. Short descriptions are 

 given of the morphology and the structure of the Cycads, the struc- 

 ture of the MeduUoseae, and the structure of Lyginodendron and 

 Heterangium. 



IL The Origin of Cycadean Structures. 



a) The Origin of Axial Structures. The view that the Cycadean 

 cylinder is derived from that of such a form as Mediillosa porosa is 

 adopted. Ancestral characters should be sought for in two principal 

 regions (1) the cotyledonary node, (2) the flowering axis. In the 

 stem of a seedling'of Encephalartos Bateri Matte has found three 

 almost independent steles, each of very sinuous contour; this struc- 

 ture is exactly comparable to that of Medullosa anglica. The irregulär 

 orientation of the bundles in the lower part of the peduncle of 

 Stangeria may be explained by supposing that the whole cylinder has 

 been derived by fragmentation of a few steles of sinuous contour 

 such as are found in Medullosa Leiickartii. If all the bundles resulting 

 from the fragmentation had remained concentric the resulting type 

 of structure would resemble M. Solmsn, in which the polystelic cha- 

 racter is probably derived from the solenostelic condition found in 

 M. porosa. The centripetal xylem found in the Upper part of the 

 peduncle is homologous with that oi Lyginodendron\ in the latter each 

 primary xylem group with its secondary xylem and phloem is the 

 one-sided remnant of an entire stele such as occurs in Medullosa. It 

 is also homologous with the Single stele of Heterangium and Mega- 

 loxylon. In some cases in L. Oldhamium concentric stele-like bundles 

 occur and this is the normal structure in Cycadoxylon: these are 

 regarded as reversions to an ancestral character. The more concen- 

 trated structure of the stele found at the base of the branches of 

 Lyginodendron is considered to be a mechanical adaptation. It is to 

 the typical mature structure and not to the early stages of its ontogeny 

 that we must look for the occurrence of ancestral characters. 



In the Medulloseae the cylindrical and medullary Systems of steles 

 are variants of a Single System; probably the former is primitive and 

 gave rise to the latter. Similarly in Macrosamia the collateral medul- 

 lary bundles are of cauline origin and constitute the same System as 

 those of the vascular ring. In the fertile part of the axis of the male 

 cone of Ceratosamia there is the rudiment of an intrafascicular pri- 



