472 HYPNACE^E. 



Var. r). tectorum Brid. In close, dense, swollen, velvety 

 cushions ; branches erect, curved, more prostrate at the outside 

 of the tufts ; dark or olive green. Leaves erect and closely 

 imbricated, very lightly curved and secund, more spreading 

 when moist ; rather wide, ovate-oblong. 



Var. 6. elatum B. & S. Robust, densely tufted, with few, 

 short, erect, very swollen branches; yellowish or golden green. 

 Leaves large, very wide and concave, and imbricated all round 

 with only the points falcato-secund, rendering the stems and 

 branches very tumid and julaceous ; widely ovate-oblong, rapidly 

 narrowed to a short oblique acumen. Fruit very rare. 



Hab. Trunks of trees, walls, earth, etc. Abundant. Fr. winter and spring. 

 Var. filiforme on trunks of trees, usually in woods, common. Var. minus on trees, 

 rare. Var. mamillatum on rocks and trees in woods. Var. ericetorum on heaths and 

 in mountainous woods, frequent. Var. tectorum on rocks, walls and roofs, frequent. 

 Var. elatum on the ground, in calcareous districts chiefly. 



The above list by no means exhausts all our forms, but it includes perhaps the 

 most important and the most striking. Though so polymorphous, H. cupressiforme 

 has a characteristic habit and a structure peculiar to itself, so that with experience it 

 is usually fairly easy to recognise it in the field, and doubtful forms will generally 

 reveal their identity under the microscope. The rather rapidly (in general) acuminate 

 leaves and the characteristic, small, opaque auricular cells are two of the prominent 

 characters of distinction. It will be more convenient to point out the difference 

 between this and the allied species under the descriptions of the latter. 



The var. resupinatum is by many authors considered a species or at least a sub- 

 species. In its extreme or typical form it is very distinct, and closely resembles 

 Pylaisia polyantha, and, when slender and barren, H. ituurvatum, the descriptions 

 of which may be consulted. But it is too inconstant in its characters, or, if it be 

 preferred, intermediate forms too closely link it with the type, to warrant this. 

 Schimper's var. longirostre is exactly the same plant as regards the fruit (which is 

 looked upon as one of the most important characters of H. resupinatum), and differs 

 only in the sub-secund, not straight and homomallous leaves. I have found this 

 variety in the same localities as the var. resupinatum ; but beyond this I have 

 frequently examined plants with the greater part of the stems having the normal, 

 falcato-secund decurrent leaves of the type, but frequently throwing out slender 

 branches with the "resupinatum" foliation; more rarely stems with the straight 

 homomallous leaves of the var. resupinatum will suddenly change their character in 

 the middle, and bear in the upper half the wide, decurved, typical "cupressiforme" 

 leaves. Some tufts with the perfectly erect, symmetrical capsules of the var. 

 resupinatum, and beaked lids half as long as the capsule, exhibit a perfectly 

 indiscriminate foliation, some branches being normal slender " cupressiforme" and 

 others distinctly ' ' resupinatum " in type. Even if all these forms be relegated to the 

 var. longiiostre Schp., it leaves H. resupinatum separated by the form and direction 

 of the leaves alone, and as this character occurs, partially at least, in the above forms 

 attributed to H. cupressiforme, it is clear that a species cannot be safely founded on 

 this single character. 



I have seen no British specimens referable to Wilson's var. minus, but I have a 

 plant collected by the Rev. A. C. Waghorne in Newfoundland which clearly belongs 

 there, and is very distinct in the recurving of the margin ; it is distinct from H. 

 revolutum in the texture and form of the leaves, and in the margin only recurved in 

 the lower half, not strongly revolute almost to apex. 



The var. filiforme is a very neat and pretty form, but all gradations may be 

 found between this and the type. Schimper includes under this head an equally 

 slender form with the leaves less regularly, often interruptedly imbricated, but it 



