0. - REVUE BRYOLOGIQUE 
submerged in a deep pool, near Birkdale, which never became 
dry, owing to ils being excavated far below the surface water of : 
the moss on which the sandhills lie, which water is known locally 
to builders as the ‘‘Ream”. This plant had no unbranched termi- 
nal portion, but was so closely and plumosely pinnate throughout, 
that Mons. Renauld at one time thought it worth recognition as 
a distinct form under the name of var. perpinnatum. 
If we extend our observation to other groups of mosses, il 
becomes apparent that something beyond light and shade helps to 
_ determine the amount of branching of the stems of mosses. In 
Rhacomitrium canescens the most pinnate stems are not the 
trailing laterally illuminated ones, but those erect and tufted 
forms represented by the var. ericoides B. et S. Many other 
mosses having an erect or tufted habit exhibit a good deal of 
lateral branching, as in Cylindrothecium concinnum, Camptothe- 
cium nitens, Hypnum filicinum, H. commutatum, H. stellatum, 
I. cristacastreusis and Æ. Schreberi. The real -cause of pinnate 
branching is probably to some extent inherent, and partly inde- 
pendent of external stimuli. 
In altempting to trace the conversion of one species to another 
there is a danger which it is difficult to avoid in this group, and 
which must not be over-looked, i e. the risk of error in naming 
badly developed specimens, suffering from abnormal conditions 
Of growth. For instance, Æ. pseudofluitans and H. Kneïffii produce 
wo analagous states, (forma gracilis, and forma laxifoha) which 
are not always easy to distinguish from each other when the 
Material is imperfect or poorly developed. Whilst therefore 
conceding the possibility of such reversion of one type to another 
as 1$ suggested by Mr Inghaim, it seems very desirable to await 
further evidence before advancing to the logical conclusion lo 
which they point. When characters show such instability as to 
change with the seasons they are scarcely sufficiently permanent 
to be of service in determining species and varieties, and can be 
__ expressed by {he term « forma », or some even lower step in the 
_ grades of classification. 
The following field observations on the forms included under 
Sanio's « Harpidia adunca », were made principally in the 
neighbourhood of Liverpool. The stations under notice are 
unfortunately somewhat limited in number, but have the 
advantage of being so near the writers residence as to have per- 
mitted frequent, oflen monthly observation, for a period of about 
fourteen years, under very varying seasonal conditions. 
: HYPNUM ADUNCUM (Hedw.) Ren. Sp. aggreg.) (non Drepanocladus 
