82 KANSAS UNIVERSITY SCIENCE BULLETIN. 



actively as a means of translocation of food materials that some 

 vestige of life would be discernible. In Salvia, for instance, 

 a healthy, vigorous condition of the suspensor cells accompanies 

 activity in food transmission. In all the species examined 

 (except Salvia), the only suspensor cells that appeared to be 

 living were those immediately adjacent to the embryo. The 

 nutrition of the embryo is accomplished through absorption 

 over its entire surface, whereas in Salvia an added source of 

 nutriment probably lies in the conducting power of the sus- 

 pensor. 



The length of the suspensor is roughly proportionate to the 

 size of the haustorium through which the embryo has to be 

 transported. Usually the movement is past a coenocytic struc- 

 ture to a tissue. An instance is to be noted, however, in My- 

 oporum^ (Myoporacese) , in which the growth is directly 

 through one portion of the endosperm to another. The beaked 

 portion of the endosperm is apparently a remnant of what may 

 once have been a coenocytic haustorium, but in which the coe- 

 nocytic condition has been lost. 



The integument varies in thickness and extent in the differ- 

 ent species. Of course, it is in those in which it is most ex- 

 tensive that the largest haustoria are possible. Through the 

 integument in each case there runs a vascular bundle which 

 terminates near the antipodal region. A slight outpocketing 

 of the embryo sac is found extending into the integument to- 

 ward the end of the bundle. Later it may develop into a canal 

 without nuclei, or into a process of the endosperm. Its func- 

 tion in all cases is evident — to facilitate the passage of nutri- 

 ment from the vascular bundle to the endosperm. 



In summing up the characteristics that are common to the 

 species described in this paper, with the exception of Salvia, 

 they have been found to possess three that in the future may 

 be shown to belong to the family as a whole. They are the 

 micropylar haustorium, the much-elongated suspensor, and the 

 antipodal canal or process. Marked variations from a common 

 type, such as are found in Salvia, suggest a taxonomic re- 

 arrangement, even though the usual macroscopic characteris- 

 tics resemble those of the family in which the varying species 

 is placed. In a previous paper,' for instance, attention was 

 called to the striking difference in the ovular structure of the 

 Menyantheae when compared with the coordinate subfamily 



