72 Salisbury .—On the Structure and Relationships of 
found with three ribs in place of two, a variation that may even be a 
reversion. 
Such examples suffice to show that the assumption of an originally 
hexamerous integument (later becoming trimerous in some forms) for the 
Trigonocarpeae does not preclude the origin of closely allied genera pos- 
sessing bilaterally symmetrical structures. (The position of the tetramerous 
Radiosperms awaits details of their internal structure.) 
In view of the frequent association in the monocotyledonous series of trimerous 
flowers with a triangular stem structure, an expression probably of similar mechanical 
relations, it is of interest to note that the general outline of the stems both of Medullosa 
anglica and Sufcliffia msignis was broadly triangular , 1 though the character does not 
of course hold for other species of the former genus. 
Before leaving this subject it is of interest to note that Drs. Stopes and Fuji 
regarded Yezostrobus Oliveri as more nearly approaching to Trigonocarpus than any 
other known fossil or recent group , 2 and therefore, on the foregoing hypothesis, the 
triangular outline which this seed exhibits in transverse section 3 may have a phylo- 
genetic significance. 
2. The Question of a homogeneous or dual Phylogeny. 
The question at once arises in relation to the integument : was this 
structure of uniform origin or was it dual, as suggested by Stopes for 
Cycads ? 4 The close agreement of the two groups makes the assumption 
of a double nature for the one almost necessitate its acceptance for the 
other. The main grounds on which this theory is based are, briefly, the 
presence of two series of bundles, both supposed to be integumental, and 
the resemblances that exist between Lagenostoma with its cupule on the one 
hand, and the Cycadean ovule with its sarcotesta on the other. The recent 
work of Miss Kershaw 5 on Bowenia spectabilis has shown that in this 
species at least the inner vascular system is nucellar. On the other hand, 
Dr. Stopes traced some members of the inner vascular system of Cycads 
into the integument beyond the free part of the nucellus, and unless future 
work should show that it is only the accessory branches from the integu- 
mental system that behave in this manner, the evidence as to the nature of 
the inner bundles must remain in its present contradictory state. But in 
view of the diverse data at present available, we are not warranted in 
attaching any great importance to the occurrence of two systems, even 
should they prove to be both integumental in certain cases. The branched 
1 Scott, Phil. Trans. Roy. Soc., B, vol. cxci, 1899; and Scott : On Sutclijffia insignis. Trans. 
Linn. Soc. Bot., vol. iii, 1906, Pt. 4, ser. 2. 
2 Phil. Trans. Roy. Soc., ser. B, vol. cci, 1909. More recently, however, Prof. Fuji has come to 
regard the nearest living relatives of this plant as being the Araucarieae. 
3 Loc. cit., Fig. 14. 
4 On the Double Nature of the Cycadean Integument. Ann. Bot., 1905. 
5 Ann. Bot., vol. xxvi, No. 103, 1912, p.636. 
