666 SUMMARY OF CCTRRBNT RESEARCHES RELATING TO 



only in a weakly condition. They grow vigorously only when they meet 

 with host-plants, especially grasses. 



The roots of BJiinanthus are but very sparsely supplied with root- 

 hairs, and consequently have but little power of independent nutrition. 

 The haustoria are formed at an early period, and are, like those of 

 Melampyrum, exogenous. Tliey penetrate the root of the hust by means of 

 a sac-like cell not unlike the apical cell of a growing-point, which breaks 

 through the endoderm of the vascular bundle of the host and enters the 

 thick-walled xylem, probably with the aid of a solvent excretion. The 

 process is somewhat different according as the host-plant is a mono- 

 cotyledon or dicotyledon. The haustoria contain no starcb, but, on the 

 other hand, minute granular or rod-like albuminous bodies, which 

 gradually fill up the tissue, but are used up by the time of flowering, 

 and closely resemble the " bacteroids " of the tubercles of the roots of 

 LeguminosEe. 



lihinanthus is, therefore, in contradistinction to Melampyrum, a true 

 parasite ; but it may also live to a certain extent saprophytically. The 

 cells of the parasite are in contact with a homogeneous yellow mass result- 

 ing from the decomposition of the cells of the host ; the vascular bundles 

 of the host being also to a large extent destroyed. From tho margin 

 of the haustorium cells also advance towards and penetrate the cortex 

 of the host and destroy it, apj)lying themselves to and embracing the 

 destroyed portion somewhat in ihe manner of root-hairs. Parasitism is, 

 therefore, necessary for the vigorous growth of Bliinanthus, but the plant 

 is, like the mistletoe, only a partial parasite, carrying on an independent 

 assimilation of its own, and not inflicting any serious injury on the host. 



M. Granel * dissents to a certain extent from the observations of 

 Leclerc du Sablon and Koch on the origin of the haustoria in Melam- 

 pyrum and Bhinanthus, inasmuch as he asserts that they do not spring 

 from the piliferous layer, but, in accordance with his previous observa- 

 tions on Orohanche and the Santalacese, from the cortical parenchyme. 

 He agrees, however, with those authorities as to the morphological 

 nature of these organs. They are of exogenous origin, not springing, 

 like the rootlets, from the pericycle, and they do not present the least 

 trace of root-cap ; the arrangement of their vascular system is also very 

 diiferent from that of roots. 



Modifications in the Roots of Grasses growing^ in Water.f — 

 M. Devaux describes certain modifications which had taken place in the 

 roots of Lolium and Holcus mollis when grown in water. In Lolium the 

 development of the roots at first appeared normal; but, after having 

 attained a length of some centimetres, a large nimiber were observed in 

 which the growing point, instead of continuing its descending direction, 

 inclined towards the side, then raised itself, and finally followed a 

 helicoid line of growth in the liquid. In Holcus mollis the root-hairs 

 preserved their normal form, but a number of tubercles were developed, 

 the radicles being arrested in their development. 



* Comptes Rendus, cviii. (18S9) pp. 367-9. 



t Bull. Soc. Bot. France, xxxvi. (1889) pp. 76-81. 



