102 Oliver . — On Physostoma elegans , Williamson , 
Length, 6 millims. ; broadest diameter, 2j millims. ; the convex faces 
of ribs and tentacles, which are usually ten in number, are densely covered 
with tubular hairs reaching a length of § millim. 
2. Physostoma Kidstonii , Arber. 
Lagenostoma Kidstonii , E. A. N. Arber, 4 On some new species of 
Lagenostomal Proc. Roy. Soc., B., vol. lxxvi, 1905, p. 245, and PL I and II. 
Length 6 millims. ; broadest diam., millims. ; ribs and apical lobes 
usually six in number. Known only as casts. 
Locality : Stonehill Colliery, Stonehouse, Lanark. 
Horizon : Lower Coal-Measures. 
IV. General Discussion. 
The Multiple Integument. 
More than one writer on Palaeobotany has drawn attention to the 
very exceptional condition presented by the free part of the integument of 
Physostoma } Comparison with such a seed as Lagenostoma Lomaxi shows 
that the united canopy finds its homologue in the circlet of free tentacles of 
Physostoma. The chambered structure of the former becomes intelligible 
when viewed in the light of the latter, and leads to the irresistible con- 
clusion — finding support in the widest possible range of analogous cases — 
that priority in this line of descent must be given to an integument of 
separate, free segments. The numerical excess of the free tentacles of 
Physostoma over the united chambers of the known Lagenostomas points 
in the same direction, i. e. to the relatively primitive condition represented 
by our seed. 2 
If we were fully informed of the range of structure of the integument 
in the seeds of the Lagenostoma-gxou p, we should doubtless recognize 
stages of coalescence intermediate between Physostoma and Lagenostoma 
Lomaxii. Such an intermediate condition may be accepted without much 
risk of error for the case of L . Kidstonii of Arber, whilst this authors 
L. Sinclairii may be a second example of the same condition. Both of 
these were ribbed seeds with notched apices, and their recognition tends to 
remove Physostoma from the very isolated position it would otherwise 
occupy. 
When we come to the question of the origin of such an integument as 
that of Physostoma, the only point that is obvious is that here, and in the 
L agenos toma-gr oup generally, this origin has been a multiple one. In the 
absence of direct evidence, the morphological nature of the unit structures 
1 Miss M. Benson, On Telangium Scottii, Annals of Botany, vol. xviii, p. 169. D. H. Scott, 
Progressus Rei Bot., I, p. 211. 
2 The usual numbers are as follows : Physostoma , ten : Lagenostoma Lomaxi , nine ; Lageno- 
stoma ovoides, eight; Conostoma oblongum , seven; Lagenostoma Kidstonii , about six. 
