10 Dr. G. Dickie on the Physiology of Fecundation in Plants, 



attached to the embryo. The ripe seed of this plant, as well as 

 other Orchidece, is quite destitute of vessels * ; in Narthecium ves- 

 sels do not appear until some time after fecundation. 



It is exceedingly probable that the emission of such prolonga- 

 tions, ovule-tubes as they may be called, to meet the descending 

 pollen-tubes, may be of frequent occurrence. It is possible that 

 some of those cases in which pollen-tubes are described as having 

 been seen entering the foramen of the ovule, may have been in- 

 stances of such prolongations as exist in Narthecium. 



Supposing it to be a settled point that contact of pollen-tubes 

 with some part or other of the ovules is necessary to secure the 

 full development of the embryo, — and some authors are of this 

 opinion, — there seem many physical difficulties in the way of the 

 pollen-tubes reaching their destination ; these have reference to 

 the length of the style, the nature of the tissue of which it con- 

 sists, the form of the ovule, the position or direction of the ex 

 ostome, the number of the ovules and other circumstances. Sure 

 it is, that difficulties cannot stand before demonstration. True 

 pollen-tubes have no doubt been in many instances traced into 

 the tissue of the stigma and style, and in some cases to the pla- 

 centa, and it may be, even to the ovule itself, but instances of 

 this last are far from numerous. In order to secure the effect of 

 the pollen upon the ovule, or at all events to diminish the chance 

 of this action failing, it may not be unreasonable to expect that 

 some such arrangement would be provided as that, the existence of 

 which in Narthecium I have been attempting to prove. Observers 

 so frequently meet evident instances of design in the structure 

 of organs, — it may even be said of less importance than those un- 

 der discussion, — that it is extremely probable the present may be 

 a case in point. 



It has been already remarked, that the number of the ovules 

 presents a difficulty in the way of the contact of a pollen-tube 

 with each ; and where they are very numerous and arranged in 

 regular series from the base to the apex of the placenta, it is very 

 evident that those occupying the latter position stand a better 

 chance of being fecundated than those toward the lower part. 

 The emission of tubes from the ovules to meet the descending 

 pollen-tubes would evidently diminish the chance of failure. In 

 such cases it would be interesting to ascertain whether all are 

 usually impregnated or what proportion of them only, and whe- 

 ther those taken from the upper part of a capsule contain a greater 

 proportion of fecundated individuals than those from the lower 

 part of the same. 



* Callitriche verna presents an instance of non-vascularity in filaments 

 and anthers. 



