384 EQUISETALES 



(Bornid) Renault describes 1 for B. radiata, Brongn., how the fructifications 

 are simple, or interrupted in their length by verticils of leaves, which 

 render the spike itself, so to speak, articulated and of very variable 

 length. The condition of these spikes is then different in proportion, 

 rather than in essential points from that described for Phyllotheca (Fig. 

 197), and so curiously reproduced in the abnormal Equiseta described 

 above (Fig. 196). This again differs from Calamostachys mainly in the 

 number of the sporangiophores which intervene between the successive 

 leaf-whorls. The tracts which bear the sporangia being thus variable, it 

 would appear that the Equisetum-ty^te is merely an extreme case, in 

 which the whole series of sporangiophores which form the terminal 

 strobilus are collectively above the last leaf-sheath, and that last leaf sheath 

 is of a reduced type, and appears as the annulus. 



It is obvious that in the present state of our knowledge the case is 

 not proved either for the phyllome-theory of the sporangiophore in the 

 Equisetales, which is out of harmony with the known facts in the fossils, 

 or for the non-phyllome theory, which is certainly a less obvious explanation 

 of the simple strobilus of Equisetum. But the balance of evidence is 

 strongly in favour of the latter, as without undue pressure it covers the 

 whole area of facts, including those relating to the fossil Equisetales. - 



1 Bassin Houiller d'Autun et d'Epinac, p. 8 1. 



2 It is necessary briefly to mention another view, advanced by Jeffrey (Mem. Boston 

 Soc. of Nat. Hist., vol. v., pp. 184-5), as applicable to those Calamitean cones where 

 the bracts in each whorl are stated to be double the number of the sporangiophores. 

 He suggests that the pairs of the sterile leaves were really dichotomously divided dorsal 

 segments of sporophylls, of which the sporangiophores were the ventral segments. It is 

 necessary to remember, however, that in the best known cones of Calamostachys the 

 bracts of successive whorls alternate, while the successive whorls of the sporangiophores, 

 considered by themselves, are strictly superposed (Scott, Progressus, p. 158) : this fact 

 appears to be fatal' to Jeffrey's suggestion, as will be obvious if the arrangement be 

 plotted out diagrammatically in one plane. It will then appear that the proposed scheme 

 would only apply to each alternate whorl of bracts, not to them all. There is also 

 against it the fact that in the Equisetales at large the arrangement of the cone with the 

 bracts approximately doubling the number of the sporangiophores is only one among 

 several different arrangements : the proposed scheme is quite inapplicable for Archaeo- 

 calamites or for Equisetum ^ and equally so for Palaeostachya (cf. Hickling, I.e.). 



Akin to Jeffrey's theory, though not coincident with it, is that of Lignier (Bull, de la 

 Soc. Linn, de Normandie, Caen, 1903, p. 162, etc.), which also is based primarily on the 

 data for the cone of Calamostachys, and upon comparisons with the Sphenophylls. His 

 view is that the sporangiophores in Calamostachys are the result of concrescence in pairs 

 of fertile lateral lobes of the leaves forming the verticil. The anatomical facts are 

 derived from Renault (Bassin Houiller et Perm. d^Autun et d^Epinac, iv. , 2, p. 130, and 

 PI. Ix.) ; the details shown in his figure, 6, of the single transverse section partially 

 depicted would accord with the theory ; but the evidence seems insufficient, and there are 

 the following positive objections to it. First, there is no structural evidence in the 

 sporangiophores themselves of Calamostachys, or in any other of the Equisetales, of the 

 presumed fusion. Secondly, in the single drawing of a complete transverse section of 

 the cone of C. Zeilleri by Renault (I.e. PI. ix., Fig. 5) there are 14 sporangiophores, but 

 only 27 sterile bracts : so that the numerical relation does not hold in the one case on 



