Flora of New Zealand Series

P. saxatilis Carse in T.N.Z.I. 59, 1929, 315.

Plant smaller in all parts, decumbent to suberect. Stipes 5-25 cm. long; lamina 10-18 × 5-15 cm.; segs deeply, acutely toothed; anastomoses fewer; reflexed protective margin very delicate.

DIST.: N., S. Lowland and coastal dry forests and rocky places from near North Cape to lat. 41º 30'. Local south of N. Auckland Peninsula. The original localities cited are: "Among detached rocks in hilly forests, Monganui County, Bay of Islands, Whangarei, Manukau County, H.C. Coromandel Peninsula, H.B. Matthews! Thames, D. Petrie!" No specimen of Carse in CM is dated in or before 1919.

 Var. pendula (Col.) Cheesem. Man. N.Z. Fl. 1906, 973. P. (Litobrochia) pendula Col. in T.N.Z.I. 20, 1888, 218. Colenso based his sp. on specimens collected on "Even shaded wet-dripping gravelly cliffs (among other ferns and shrubs), banks of a stream south of Danneverke Township, County of Waipawa; 1887: W.C." and gives a long diagnosis and discussion. "Fronds pendulous, broadly deltoid, 10-12 (rarely 14) inches long . . . ultimate pinna 3-5 inches, narrow-ovate or lanceolate-acuminate; segments very distant and decurrent . . . large, open . . . their margins and apices coarsely and deeply incised . . . stipes 7-9 inches long . . . scales dark brown, subulate . . . very acuminate . . . Sori in short narrow lines (and dots) . . . not at sinus or near tips." Cheeseman's diagnosis is: "not so finely divided. Terminal pinna larger, 3-5 in. long, often caudate. Pinnules larger, 2-21/2 in. long, ovate, acuminate; segments longer and narrower."

I have not found Colenso's type. There is a specimen in W, without locality, labelled (by Colenso) "Litobrochia pendula Col. (small variety)". Stipes up to 10 cm. long, lower half with scattered ovate- to linear-attenuate paleae. Pinnae distant, subopp., lower ones stalked, up to 9 × 6 cm. Sori and segs much as described by Colenso. Forms ± agreeing with either Colenso's or Cheeseman's description occur throughout the range of the sp.

POLYMORPHY

The Pteris complex in N.Z. is still very imperfectly resolved, and the relation of our forms to overseas ones needs fuller study. Cockayne and Allan (Ann. Bot., Lond. 48, 1934, 9) remark: "Carse, who had given a good deal of study to it, viewed the three species recognized by Cheeseman as linneons each composed of several jordanons, any of which appeared to hybridize freely when meeting with another. Our own observations are in accord with Carse's." They record P. comans × macilenta and P. macilenta × tremula, but these crosses cannot be regarded as definitely established.