erable numbers in the diseased tissues ; and since the 

 presence of a sinsjle one of these bacteria may cause the 

 disease, we should guard against dangers of contamination. 

 Although it has never been demonstrated, yet it seems 

 probable that the bacteria present in the diseased seed, lint 

 and carpels, after they fall to the ground and become disen- 

 tegrated, are elibtrated and find their way to the roots of 

 the cotton plant which they enter, and pass up through the 

 pliant to the V)olls, inside of which they find conditious suit- 

 able for their development. Or the seed may lie unaffected 

 but the lint left attached to it may contain the bacteria 

 which would thus be in close connection with the young- 

 cotton plant when it germinates, and then could find its way 

 into the roots. And it also seems very probable that those 

 seed which are affected with the bacteria, but not in suffi- 

 cient quantities to prevent their germination, may produce 

 young plants with the rot bacteria already within their 

 tissues (seed leaves), and thus these bacteria may then easily 

 find their way into the bolls when they appear. But it 

 seems to me even more probable that the bacteria are carried 

 by the wind or insects from the soil to the flowers, where 

 they remain attached to the moist and viscid stigma or in 

 the nectar ; and that they not only thus readily find their 

 way into the young and developing bolls, but that they even 

 multiply in the nectar or on the stigma; and that the insects 

 which visit the flowers are thus contaminated and inoculate 

 other flowers. This seems even more probable since we 

 know of certain other bacterial diseases of plants, as pear 

 blight, that is thus carried from one tree to another, and 

 from one flower to another on the same tree. This exjDlau- 

 ation of the spread of the disease helps us over one diffi- 

 culty, namely, the fact that the disease is principally con- 

 fined tp the middle and top crop. For if the bacteria are 

 in the young cotton plant before the bolls are formed, one 

 would expect the first or lower crop to be equally affected. 

 If the bacteria enter hj way of the flowers, we could ex- 

 plain the scarcity of the disease in the lower or first crop 

 of bolls by the supposition, that the insect which carries 

 the disease from one flower to another does not appear until 



