Mesoxylon Lomaxii and M r poroxyloides. 1025 
strands of primary xylem surround the pith, most of them in contact with 
the secondary wood, though some are separated from it by 1 to 3 rows 
of cells. These strands anastomose freely. At a distance from the pith the 
rays are usually uniseriate. The pith is horizontally ruptured (imperfectly 
discoid), and the pith-cells surrounding a primary xylem-strand are elon- 
gated radially with respect to it. Tangential pits occur in the secondary 
wood, as in Pitys antiqua. The agreement with Pitys seems to be 
decidedly nearer than with the other genera ( Parapitys and Eristophyton ) 
with which the author compares it. He separates the plant generically on 
the ground of its narrow medullary rays, the wedge-like segments of secon- 
dary wood, and the arrangement of the pits, on the radial walls of the 
secondary tracheides, in groups (Zalessky, ’ll 1 , p. 28). At any rate this 
new form, of which a fuller description is promised by the discoverer, finds 
its place in the family Pityeae, and has no near affinity with Mesoxylon. 
The fragment named Caenoxylon Scotti by Dr. Zalessky is of uncertain 
origin ; it is of Upper Palaeozoic age, and possibly comes from the Permian 
of the Ural (Zalessky, ’ll 2 ). The pith is 2 cm. in diameter, and the 
zone of wood, so far as preserved, about 13 mm. thick. The presence 
of distinct annual rings in the wood is an interesting feature, indicating 
a relatively late age. The pith is not stated to be discoid, but is described 
as consisting of distinct outer and inner regions, with a meristematic layer 
between them. 
The primary xylem is divided into a number of bundles of various 
sizes and shapes, separated by medullary tissue with occasional signs 
of meristematic division. Some of the strands may be a considerable 
distance apart from the main mass of primary xylem, of which they are 
ramifications ; in all, the structure is endarch. The leaf-traces are double, 
and the medullary rays uniseriate. The author regards Caenoxylon as 
undoubtedly allied to Eristophyton , Pitys , Parapitys , and Mesopitys , and 
more remotely to Cordaites , Poroxylon , and Mesoxylon. He suggests, on 
the ground of the arrangement of the primary xylem and the character 
of the ^double leaf-trace, that the fossil may possibly be on the line of 
descent of Ginkgo. 
At present we have only a preliminary communication from Dr. Zales- 
sky on this form. Except that I cannot recognize any but the most remote 
affinity to his Eristophyton , I have nothing to add to the author’s remarks. 
In connexion with Caenoxylon we may shortly refer to Dr. Zalessky’s 
new genus Mesopitys , founded on the Araucarites Tchihatchejji of Goppert. 
This, again, is a Permian plant, and was the first stem of that age in which 
the presence of annual rings in the wood was observed. The pith is small 
(3 mm. in diameter in one case), the stems (without cortex) ranging from 
5x3 to 13x8 cm. No mesarch xylem-strands were observed ; the primary 
wood, as in Caenoxylon , is all centrifugal, forming groups of irregularly 
