358 NATURAL SCIENCE. May. 



to are very explicit, and leave no doubt that the former alternative is- 

 the true one. 



In the next place, it became clear that the axis differed in a 

 much more important point from the published descriptions of it, 

 and that in a direction which had not been previously suspected. In 

 fact, the specimens proved to demonstration that the cellular pith is- 

 surrounded by a ring of primary vascular bundles, which, in essen- 

 tials, are of the true Calamitean type. The number of these is- 

 usually three, though there may be four or more, but, whatever be 

 the number, each is characterised by the presence, at the apex of the 

 xylem, of a carinal canal homologous with the canals found in the 

 same position in the stems of Calamites and Eqiiisetum, and which 

 represent the tracheal initial strands of the primary vascular bundles. 

 In the transverse sections, we find individual elements clinging to- 

 the sides of the canals, and projecting into their cavities, reminding 

 us at once and forcibly of the appearances met with in Eqiiisetum. SO' 

 close is the resemblance, that one might from the transverse sections 

 alone infer that these projecting elements are annular or spiral 

 vessels, but for demonstrative proof recourse must be had to the 

 longitudinal ones. Fortunately, some of these have been made in 

 such favourable directions, and the tissues are so well preserved, that 

 they leave nothing to be desired. They show us the carinal canals cut 

 throughout their whole length, with the annular vessels still attached 

 to the peripheral walls. 



Hence it is obvious that the stele or central cylinder of the axis 

 of Calamostachys Binncyana has a structure which is identical in essen- 

 tials with that found in the stem of Calamites, and there is no such 

 antagonism between the anatomy of the two as has hitherto been 

 believed. Not only so, but when, as sometimes occurs, the forma- 

 tion of secondary xylem begins, it does so first at the carinal canals,, 

 as in Calamites, and subsequently in the intervening areas. In this 

 way is brought about the triangular form of the pith, and the 

 triquetrous arrangement of the vascular laminae which surround it, 

 and which Williamson compared with the secondary xylem of the 

 plant he named Astcrophyllites. 



The conclusion, then, is, that the correspondence between the 

 anatomy of the stem of Calamites and the axis of Calamostachys 

 Biuneyaiia is so close and definite as to leave no doubt whatever 

 that the one is the fruit-spike of the other. Williamson's objections 

 to this determination have all been based upon the supposed 

 absence of such a correspondence, and now that it is shown to 

 exist they naturally fall to the ground. The internal anatomy of the 

 spike, then, points to the same affinities as the external characters, and 

 together they supply all the evidence that is needed to establish the 

 conclusion we have stated. To which of the forms of Cfl/^wZ/^s the spike 

 is to be more particularly referred cannot at present be definitely 

 stated, but there are some reasons for thinking that it must be sought 



