PROSOPIS VELUTINA; ZIZYPHUS PARRYI. 19 



The roots of Prosopis in which haustoria wpre found were 1.5 mm. to 1 

 cm. in diameter. The diameter of the roots of the parasite just above or 

 between the haustoria was usually about equal that of mesquite, but it 

 might be less, as in one instance a mesquite root 5 mm. in diameter bore 

 a haustorium whose diameter was only 1 mm. 



The repeated formation of haustoria by a single root of Krameria was 

 most noticeable in its connection with Prosopis. In one case 10 haustoria 

 were given off within a space of 8 cm., and in another instance the haus- 

 toria were 1.25 cm. apart. Usually the haustoria were sessile, but they 

 also terminated short roots. From the fact that the haustoria frequently 

 arose directly from the side of the root, the portion of the parasite root 

 between the haustoria usually comes into contact with the host root. Con- 

 tact alone, therefore, of old tissues does not bring about the formation of 

 new haustoria; they are formed in young parts exclusively. 



A transverse section of the host root, 9 mm. in diameter, with an haus- 

 torium attached (figure 6, plate 4) shows the haustorium penetrating 

 slightly into the woody cylinder. The leading points of interest are in 

 relation to the position of the haustorium in the cortex and its relation to 

 certain tissues there. In the cortex of the root of Prosopis are to be found 

 several concentric rings of hard bast, apparently continuous, and which 

 therefore would be an effectual barrier against the encroachments of the 

 sinker of the haustorium if it were not able to absorb them. It appears 

 from examination that the haustorium is able to do this to a considerable 

 degree. This conclusion is based on two observed facts: (l) that the 

 sinker has penetrated to the central cylinder of the host, and (2) that there 

 are no hard-bast groups to be seen in the haustoria, which would be the 

 case if the lignified elements were merely engulfed. The first observa- 

 tion would be meaningless if it were the question only of the primitive 

 portion of the haustorium, which might have penetrated during the earliest 

 stages of development of the host root, when the tissues of the cortex were 

 yet undifferentiated. But after penetration was effected the haustorium 

 extended some distance into the cortex, and in doing so broke down the lig- 

 nified tissues which stood in the way. (Fig. 7, plate 4.) 



ZIZYPHUS PARRYI. 



The habitat of Zizyphus, in the vicinity of Tucson, agrees very exactly 

 with that of Prosopis velutina. It inhabits the washes and flats and attains 

 its largest size on the flood-plain of the Santa Cruz River. 



The root-system is a deeply penetrating one. On the flood-plain there 

 is developed a stout tap-root which goes down 2 meters, and probably 

 farther, and a few laterals which do not extend widely away from the main 

 axis. Zizyphus is one of the desert plants with a pronounced xerophytic 

 aspect, but which nevertheless must have a relatively copious water-supply 

 throughout the year. It is one of the plants also with the character of 



