1903.] NATURAL SCIENCES OF PHILADELPHIA. -681 



21. Viola pedata lineariloba D. C. 



Viola pedata lineariloba D. C, 1824, Prodr., I, p. 291. 

 Wiola pedata rammculi folia D. C, 1824, Prodr., I, p. 291. 

 Viola pedata inornata Greene, 1896, Pittonia, III, p. 35. 



iSangre.— Southeastern counties of Pennsylvania and in southern New 

 Jersey (apparently not in the typical pine barrens). To the north- 

 ward it ranges to Monroe, Northampton, Lancaster and Huntingdon 

 counties in Pennsylvania (Porter), and to several stations in the north- 

 ern counties of New Jersey— Sussex, Essex, Union, Somerset, Warren 

 and Himterdon (Britton). 



Habitat.— Dry sandy banks, or sandy open woods and clearings, 

 necessarily local, though often very abundant. The nearest stations 

 to Philadelphia are the serpentine outcrops of Delaware county and 

 sandy pine woods near Springdale, New Jersey. 



Description.— Eavly flowering plant, Media, Delaware county, Penn- 

 sylvania, April 26, 1903. No. 5,117, Herb., W. S. Flowers large, flat 

 —i.e., petals nearly in one plane~20-35 mm. in diameter, all glabrous, 

 and without darker lines, blue (or occasionally lilac or white), stigma 

 not bearded nor rostrate; sepals long, linear-lanceolate; scapes 

 glabrous, 70-90 mm. (stunted plants in very barren soil have them 

 much shorter). I>eaves glabrous, all divided into 7 to 11 linear lobes, 

 toothed at the end ; the earliest leaves have the lobes less numerous 

 and spatulate; size 15 x 18 mm.; later leaves 25 x 25 mm.; petioles 

 20-50 mm., glabrous. 



Late fruiting plants. Media, June 4. No. 5,118, Herb., W. S. 

 Leaves 45 x 50, petioles 100 mm. long. Fruiting scapes 80 mm. No 

 cleistogenes are produced in this species. 



The variety of size, both in plants and flowers, owing to the nature of 

 the soil, is very great. I have specimens in full bloom in which the 

 leaves are 2 inches high and the flowers 3, while from another locality 

 I have plants of apparently the same age in which the leaves reach six 

 inches and the flowers 7. Some anomalous plants collected in late 

 summer have very broad segments to the leaf — in fact, they are some- 

 times reduced to coarse teeth— and are minutely pubescent. These I 

 take to be a second crop of leaves, produced through some unusual ac- 

 tivity, or perhaps due to arrested development in spring, since the 

 leaves usually wither away by midsummer. Such plants have, more- 

 over, been collected in full flower in late August and September. The 

 lobing of Viola pedata and V. p. lineariloba is on a different plan from 

 that exhibited in the foregoing cut-leaved species. In all palmate or 

 pedate violets the primary division is in three parts. In V. palmata dila- 

 tata the process frequently stops at the trilobed stage, but in V. scp- 



