On the Ocewrrence of Arthrostigma gracile. 105 
published in 1882, Dawson supplements the description 
of the fruit of Arthrostigma from specimens found at 
Campbellton. He says: “At Campbellton also the cones 
are better preserved, and I have figured one of them on 
plate xxiv. fig. 22. They have apparently been cylindrical, 
but there seems reason to doubt whether they were strobiles 
bearing very thick and somewhat open scales, or spikes of 
sack-like spore-cases. The Campbellton specimens certainly 
favour the latter conclusion, and if this is correct, the fructi- 
fication of this plant was of a very peculiar character, and in 
some respects more nearly allied to that of Pstlophyton than 
to that of true Lycopods.” 
“From these additional specimens, Arthrostigma gracile 
would seem to have been a small shrubby plant, with stems 
not exceeding an inch in diameter, and sparsely covered with 
conical spine-like leaves, which left, when detached, round 
scars like those of Cyclostigma. The branches, which were 
developed by bifurcation, were densely crowded with acicular 
leaves nearly at right angles to them, and were terminated 
by cylindrical spikes of fructification.” ! 
The genus Arthrostigma has been compared by some with 
Haughton’s Cyclostigma,? and they appear to agree in both 
having circular or oval leaf-scars, but with this single point 
of resemblance, any further comparison ceases. Instead of 
the abortive thorn-like leaves of Arthrostigma, Cyclostigma 
has long well-developed grass-like foliage, and the fruit is, 
I believe, the cones with long linear bracts, described by 
Schimper as Lepidostrobus Batlyanus.? Cyclostugma further 
attained to arborescent dimensions. In Arthrostigma the 
leaves are rudimentary and spine-like, and the fruit consists 
of small cone-like structures, most probably consisting of a 
spike of sack-like spore-cases. Dawson believes the cones 
were probably cylindrical, but if one may judge from his 
figures, the sporangia appear to be ranged in two opposite 
rows. But whatever may have been the minute structure of 
the cone, Arthrostigma is essentially distinct from Cyclostigma 
1 Loe. cit., p. 104. 
2 Read Roy. Dablin Soc., May 27, 1859. Jour. Roy. Dublin Soc., vol. ii. 
3 Traité d. paléont. végét., vol. ii., p. 71, pl. Ixi., figs. 9, 9a, 90. 
