ternately and then pinnately decompound, very thin and deH- 

 cate. caraway-like, the final divisions minute and fihform- 

 oblanceolate, 1-2 lines long- and flat ; the stems are erect, or 

 nearly so, and with about 3 nodes ; the leaves like the rest of 

 the group have no proper petiole, the much dilated sheath 

 reaching- to the blade ; it is a slender plant with slender rays 

 verv unequal, the longest not over 3 inches long; the ravlets or 

 pedicels are not as long- as the fruit, but filiform. The ribs 

 are conspicuous and raised. I doubt if this is found in Cali- 

 fornia, It is manifest that Coulter and Rose have referred 

 much southern California material to this species, which be- 

 longs to the next. 



C. caruifolia is well characterized by Torrey and Gray 

 and by Jepson. It is nearly acaulescent with several long and 

 stout peduncles from the same root ; leaves ternately and then 

 pinnately decompound, the ultimate segments in the type 

 linear to narrowly oblong, 2-10 lines long, not thin and deli- 

 cate; fruit 3-6 lines long, broadly elliptical, the wing: about 

 half the body, oil tubes very obscure or none on the face and 

 none on the commissure in the type. This includes Peuceda- 

 num erosum Jepson. Apparently confined mostly to the San 

 Francisco region, though found as far south as Bakersfield. 



C. caruifolia van patens n. var. 



This variety includes the normal forms with linear- 

 oblanceolate seg-ments 2-6 lines long, normal fruit 3-4 lines 

 long with wing not wider than body, and evident oil tubes, 

 those on the face occupying the entire space between the ribs 

 and seeming single but really 2 to 4 in the intervals and 4-6 

 on the commissure. The type of this is my specimens from 

 Tehachapi. California, May 20th, 1903. Probably the common 

 form of central California. 



C. caruifolia var. Vaseyi (C. & R. Bot. Gaz. 13 144 (1888) 

 as Feucedanum). 



This is characterized by the obovate segments, fruit with 

 wings wider than the body and nearly 6 lines long, and by the 

 occasional absence of oil tubes on the commissure. This is 

 the prevailing form of southern California from Tehachapi 

 southward. All sorts of intergrades are found between this 

 and the tvoe and var. patens. The pubescence is so variable 

 as to be of little value. The last variety is reported as far east 

 as Darwin mesa in the Death Valley region. I have ntatenal 

 of what is probably another species of this group from the 



