10-JO 



part of the bracken used as a thatch, and a very durable thatch the 

 stipites of the bracken do form, can scarcely doubt that the stipes of 

 Sphenopteris would have served the purpose equally well. Evidently, 

 were it still in existence to be employed for that purpose, a roof 

 thatched with Sphenopteris, with its pinnse and leaflets concealed, 

 and only its club-like stems exposed row above row, in the ordinary 

 style of the fern-thatcher, could not be distinguished, so far as form 

 and size went, from a roof thatched with Pteris. At a height of from 

 seven to eight inches above its club-like termination, the stem divided 

 into two equal parts, which shot upwards with a divergence that ren- 

 dered the fork between an angle of about 30° ; and at unequal heights, 

 a little further up, each of these divided stems bifurcated, in turn, at 

 about the same angle, and then shot up, in some individuals, without 

 further bifurcation ; while in others they bifurcated again, and yet 

 again. It is probable that, as in many of the recent ferns, the greater 

 divisions of the plant were constant, while the smaller vai'ied accord- 

 ing to the richness of the soil, and the consequent size and degree of 

 development attained by the frond. As in Pteris aquilina, there shot 

 out from these main stems numerous pinnae irregularly alternate, and 

 which, becoming less compound as they approached the top of the 

 plant, passed, in ascending, from tripinnate to bipinnate, and assumed 

 finally the form of more alternately pinnate leaflets. Unlike Pteris, 

 however, whose stem remains bare of pinnee until its larger divisions 

 take place, the stem of Sphenopteris elegans sent forth on its oppo- 

 site sides two decompound pinnae, the one about an inch, the other 

 about an inch and a quarter or so, below the first fork, — a peculiarity 

 of structure that must have imparted a graceful fulness of outline to 

 the lower portion of the frond, which, had the rachis been bare, it 

 could not have possessed. Alternation, save in the bifurcations of 

 the main, secondary, and tertiary stems, and in the case of a few irre- 

 gular pinnae that seem to have been placed opposite, or nearly so, 

 constituted the law that regulated the form of the plant. The pinnae 

 alternated on the greater stems, the semipinnae alternated on the 

 pinnae, and, finally, the minute, closely nerved, spathulate leaflets 

 alternated on the semipinnae. The entire frond must have been of 

 great lightness and beauty, of a style intermediate, from the slimness 

 of its leaflets and the slenderness of its secondary and tertiary stems, 

 between that of the frond of Pteris aquilina, and that of the fully 

 developed sucker of the graceful Asparagus. A hill-side clothed with 

 these delicately fronded ferns must have rolled its mimic waves of soft 

 green to every light breeze that stirred the depths of the old carboni- 



