GYMN0SPERAL11. 97 



was greater laterally than vertically. The remains of the Conifer are occasionally 

 associated with reedy plants, but are more often found almost unmixed with other 

 vegetation. 



Podocarpus Campbelli, sp. nov. Plate XXVI. 



Basaltic Formation ; Ardtun Head, Isle of Mull. 



The leaves are linear-lanceolate, straight or slightly curved, tapering gradually from 

 near the base to a sharp apex, constricted at the base, slightly twisted at the point of 

 attachment, and decurrent as in all distichous Conifers. Their extreme length does not 

 exceed 7 centimetres and breadth 7 millimetres. The mid-rib is distinct, but not salient. 

 There is no trace of secondary ribs, but the whole surface has a finely striate and silky 

 texture. The leaf was leathery and the epidermis dense, with the peculiar loosely-fitting 

 appearance seen in some recent Podocarps, such as P. neriifolia. Its thickness and the 

 peculiarity of its structure is revealed where one, and even two distinct layers of tissue 

 have peeled off without injury to the contour of the leaf. The margins, as in all dense 

 and leathery leaves, are particularly well-defined, and were, perhaps, very slightly 

 reflected or curled, as in some living species. The very thin outer epidermis is of a pale 

 vandyke-brown, but when this is stript, the substance of the leaf exposed underneath is 

 warm drab in colour. The specimen, Plate XXVI, fig. 3, shows a longitudinal stripe of 

 richer brown on each side of the mid-rib. The leaves were generally shed singly, and 

 are found scattered on the surfaces of the stone as in fig. 1, where the rupture is 

 seen to have been transverse, leaving the detached leaf without petiole, and with a con- 

 stricted and truncated base. More rarely they fell still attached to the branchlets, as in 

 figs. 2 and 3, when they are seen to have been sparsely and roughly arranged or pressed 

 into two rows, though somewhat irrregularly, being neither opposite nor alternate, and 

 generally set at almost right-angles to the stem. It seems that this irregular spacing was 

 produced through some of the leaves never becoming developed beyond the condition of 

 closely adherent scales, as in the recent P. neriifolia, especially when individual leaves 

 were large. The branchlets were short, truncated at the apex, with each year's growth 

 marked off by an encircling scar or cicatrix, giving the stem externally a jointed appear- 

 ance, common to almost all the living species of Podocarp belonging to the same 

 group. The cicatrix is superficial, and leaves no mark of its presence in the interior of 

 the wood, so that the branchlets do not always break up where they occur. The leaf- 

 scars left on the stem are very ill defined, and the older bark merely looks stringy. 

 The terminal bud (fig. 2) is relatively small. A complete shoot with only three 

 leaves is represented on fig. 1. Those on fig. 3 are unusually small, and I should 

 say from an examination of the dried specimens in the Kew Herbarium, that the 



