146 SCULPTURE OF THE TESTA OF GERMAN JUNCACEJE. 
ception of the common North American J. tenuis, Willd., in which the 
longitudinal ribs are so subdued that they disappear altogether on 
maturity,* and which exhibit on the surface closed and sunken meshes) 
than to distinguish them as “ semina transverse reticulata ” from those 
in which the meshes are nearly of equal dimensions. We have the 
very appropriate term “ coste " for the longitudinal ribs or lines, but 
have to select another for the transverse lines, “ /ineolg"" being used 
by Engelmann, not only for the raised transverse lines forming the 
meshes, but also for those transverse lines which terminate within the 
meshes, so common in the Junci, with articulated leaves. I shall 
therefore in the remainder of the paper employ the term “ transtilla,” 
for the raised transverse lines forming meshes; and for the more 
delicate ones within the meshes, that of * Zineole." 
now add an enumeration of the German species of Juncus, with 
parallel diagnoses of their seeds; but in submitting this I must make 
a few remarks. The descriptions refer, as do those of Engelmann's 
paper, to the external aspect of the seeds in a dry state, taken from 
herbarium specimens; they do, therefore, not claim to supply the 
anatomical details, the real cause of these differences of sculpture. 
Such details, which can only be gathered from living specimens, must 
be reserved for future investigation ; it is only with regard to the long 
caudate seeds that I have pointed out the great difference of the inner 
and outer testa, as in these it is of special importance. The observa- 
tions are best made with reflected light, and a power of 50x of a 
compound microscope: a simple microscope is less suited for looking 
over a large field. It has also to be borne in mind that the state- 
ments refer to seeds as ripe as possible. Seeds quite ripe are, how- 
ever, not very plentiful, especially in alpine species, though most 
specimens for the herbarium are collected in fruit. Quite unripe seeds 
are flat or folded (but on account of the early hardening of the outer 
membrane but slightly shrivelled), sometimes resembling the well- 
known pockets (taschen) in Prunus domestica, Half-ripe seeds retain 
a more or less cylindrical form, but they are folded longitudinally, oT; 
at all events, their ribs are much more prominent than at maturity. 
This point has to be well attended to, one being always inclined " 
classify such seeds with the “seminibus costatis ;” and, to avoid 
TI d not possess amongst my numerous specimens a single one with seeds 
e ripe. 
quite 
