Manchester Memoirs, Vol. Iv. ( 1 9 1 1 ), A^^. !K 9 



the reo^ion in which the small basal spike is given off. 

 Here the bundles on the side nearer the spike divide up 

 further until there are fourteen bundles shown in transverse 

 section {Text-fig. V., i). Those destined to supply the 

 lateral spike now bend round and are nipped off gradually 

 from the main system to form an accessory one {Text-fig. 

 v., 2, 3). The small spike was cut off just above its base, 

 and found, on sectioning, to exhibit a structure identical 

 with that of a typical specimen. 



A comparison of Bower's figures illustrating the 

 supply to the lowest fertile spike in Ophioglossum 

 palmatum, and the first three diagrams in Text-fig. V. will 

 illustrate the striking parallelism exhibited by the two 

 cases. Sections further up the stem revealed the fact that 

 the subsequently produced structures {Text-fig. II. B, j,, ^-4) 

 also conformed in a perfectly clear manner to this type of 

 branching, differing only in detail, whilst the portion s 

 was found to be equivalent to the upper portion of the 

 normal Ophioglossum spike, representing the continuation 

 of the main axis. It therefore appears that the specimen 

 described represents a single fertile spike, which, by a 

 process of chorisis, has given rise to the branched condition 

 typically found in O. pahnatmn. 



A further point in support of this view is afforded by 

 the perfectly normal character of the vascular supply to 

 the parallel series of sporangia. Had these been the 

 result of the fusion of two independent axes, either the 

 vascular strands of these axes must have rotated correla- 

 tively with the sporangia to produce their vascular supply, 

 or, in the event of there being no rotation, those sporangia 

 on the side remote from the parent axis would have been 

 devoid of vascular supply. 



A careful examination of the whole series has con- 

 vinced me that there is a perfectly normal vascular supply, 



