48 MASS. EXPERIMENT STATION BULLETIN 202. 



several species of Antirrliinuin and with several species of Linaria, including 

 Linaria vulgaris ]Mill. ' He obtained no infection on any species of Linaria. 

 The only species of Antirrhinum except the cultivated snapdragon (.4. 

 ma jus) upon which he obtained infection was A. maurandioides Gray, in 

 which case a few urediniospores were produced. 



The writer made several attempts to infect Linaria vidgaris !Mill. and 

 L. Cymbala7-ia (L) Mill, by spraying the plants with water containing the 

 urediniospores and placing them at the optimum temperature for their 

 germination (10° C). There was no infection whatever, although check 

 plants of snapdragon similarly treated became badly rusted. It is unlikely 

 that the rust occurs, at least in New England, on any other host plant 

 than the cultivated snapdragon. 



Disseinination. 



Many growers take their cuttings at a time when snapdragon rust is at 

 its height in the greenhouse, which is in March, April and May. There are 

 various opinions as to the relative merits of raising plants from seeds and 

 from cuttings, but plants raised from cuttings come true to color, and in the 

 effort to preserve a good variety some growers take rust-free cuttings 

 from a bench showing rust, or they take cuttings bearing spore pustules. 

 Somp growers have thought that if the cuttings show no spore pustules 

 they are safe to use, even though they come from a bench of infected 

 plants. The WTiter propagated plants by cuttings which bore uredinia 

 and by cuttings which bore no uredinia, although taken from an infected 

 bench. After three weeks all of the cuttings which bore uredinia were 

 badly rusted, and 35 per cent of the cuttings which when made were 

 apparently free from disease, although taken from an infected bench, 

 showed the disease. It is evident, then, that cuttings bearing s})ore 

 pustules may be expected to develop into rusted plants, and that cuttings 

 free from spore pustules, if taken from a bench of infected plants, serve 

 to aid very materially in the dissemination of the fungus. INIicroscopic 

 examination of the leaf surfaces of these apparently healthy cuttings 

 revealed numerous urediniospores which had fallen there or had been 

 carried there from diseased plants. These spores need only the favorable 

 environment of the cutting bench to cause them to germinate and infect 

 the young jDlants, and it is therefore usually inadvisable to take cuttings 

 from a house showing rust. Both Peltier {loc. cit.) and Stone (1917) 

 consider cuttings as being among the principal means of dissemination. 



Proof of the importance of greenhouse insects in the spread of snap- 

 dragon rust is not lacking. Three insects often found on snapdragon are 

 the white fl}' {Aleyrodes vaporariorum West), the red spider {Tetrani/chus 

 himactdaius Harvey) and the common aphid (Aphis gossypii). With a 

 binocular microscope the writer examined these insects on snapdragon 

 foliage. They were on healthy plants, but there were rusted plants in 

 the same bench. On the bodies of most of the insects examined were the 

 urediniospores of P. Antirrhini. 



