346 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1936 



ticle-size of tobacco necrosis virus is only 20-30 millimicrons, and that 

 of a newly described tomato virus is only 17-25 millimicrons. Again, 

 there is the recent claim of Dr. Stanley [14]^ of the Rockefeller In- 

 stitute in Princeton that he has succeeded in producing a crystalline 

 protein which has the properties of the virus of tobacco mosaic. 

 This he considers to be an autocatalytic protein, i. e., one which 

 acts upon the cells of the host in such a way as to compel them to 

 produce more of the same substance. 



For the present it will perhaps suffice to adopt the definition of 

 viruses given by Gardner [5] — "as agents below or on the border- 

 line of microscopical visibility which cause disturbance of the func- 

 tion of living cells and are regenerated in the process." 



In this short survey of the plant virus problem, it will be possible 

 to deal only with one or two of the more interesting aspects of the 

 subject, and it is proposed, first of all, to discuss a few of the symp- 

 toms produced in affected plants. Since the pathological effect on 

 the plant is almost the only criterion of the existence of a plant virus, 

 the study of symptoms necessarily plays rather a large part. There 

 are various kinds of virus diseases which may be loosely grouped 

 together as follows, the mosaic type, where attack on the chlorophyll 

 induces the formation of mottlings or rings (see pi. 1, fig, 1) ; the 

 destructive type, which induces necroses of the cells in leaves and 

 stems, and a third type which produces deformities or overgrowth in 

 the affected plants. 



Some of the mosaic viruses produce color changes in the flowers of 

 affected plants. Perhaps the best known example of this phenomenon 

 is the so-called "tulip breaking", in which tulips affected with a 

 mosaic virus produce variegated flowers (pi. 1, fig. 2). Certain of 

 these tulips with variegated flowers at one time fetched large sums 

 of money owing to the mistaken idea that they were new varieties, 

 whereas they were in reality only diseased specimens of self-colored 

 varieties. References to this tulip "breaking" may be found in the 

 literature of very early times. Thus, the first record is a description 

 published in 1576, and other accounts of this variegation in tulips 

 appeared in 1622 and 1670. It was in this latter account that the sug- 

 gestion was first made that the variegated tulip might be diseased. 

 In the Rembrandt exhibition recently held in Amsterdam were paint- 

 ings of tulips by Dutch artists of the sixteenth and seventeenth cen- 

 turies, and many of these tulips showed a typical mosaic infection. 

 Just recently, growers of the favorite blood-red variety of wall- 

 flower have been perturbed by the appearance of an ugly yellow 

 stripe or flecking in the red flowers and this has led to many com- 

 plaints from customers that their color schemes have been spoiled; 



* Numbers in brackets refer to list of references at end of article. 



