1922J 



PAYSON — STUDY OF THELYPODIUM AND ITS IMMEDIATE ALLIES 247 



typic genus. The plant in question differs from all species of 

 Thelypodium in the toothed petals and in the uniform septum. 

 It has evidently given rise to no species now included in Thely- 



dium nor has it been derived 



If 



it were placed in Thelypodium it would introduce an anomalous 

 element into an otherwise homogeneous group. All things con- 

 sidered then, it seems best to retain Chlorocrambe as a unit dis- 

 tinct from either Thelypodium or Caulanthus but probably more 

 closely related to the former than to the latter. 



CAULANTHUS 



The gynophore.—A slender stipe of one millimeter or more is 

 not known in any species of Caulanthus. In some species, how- 

 ever, the pods are not quite sessile upon the torus. This short 

 stipe, if such it may be called, is always thick and never com- 

 parable to that found in many species of Thelypodium. There 

 is obviously but one conclusion to be drawn from this: if the 

 long stipe is held to be a primitive character, then as a genus, 

 Caulanthus is not so primitive as Thelypodium. 



_ The pods.— In Thelypodium it was pointed out that those spe- 

 cies which possessed long pods had preceded those with short pods 

 in the sequence of development. An analysis of the length of 

 the pods in Caulanthus shows a high degree of intra- specific vari- 

 ation and only in a very general way may the species be compared 

 on this character. Because of the occurrence of long pods in 

 species otherwise primitive and not because of any a priori argu- 

 ment it is believed that in Caulanthus, as in Thelypodium, long 

 pods have preceded short ones in point of time. 



With two exceptions the pods of this genus are practically 

 terete. In C. heterophyllus they are somewhat compressed or 

 quadrangular and in C. calif ornicus the valves are slightly keeled. 

 These species are evidently not closely related to one another nor 

 may any developmental series be made by considering them prim- 

 itive. They are obviously recent and more or less aberrant mem- 

 bers of the semis. 



Styl 



majority of the species the 



very short and surmounted by a large stigma. The length of the 

 style seems not to be constant for the species. One group of 

 species, however, is characterized, in part, by the possession of 

 a style that tapers appreciably from base to apex. This type 

 has been described as "conic" in the key. This character reaches 



