60 THE AMERICAN BOTANIST 



testing in which the refuse was destroyed instead of being 

 dumped in a pile to spread possible fungus infection. No 

 canna plants were grown on the place ,with the exception of 

 the two mentioned, from seeds found in 1912 and none were 

 grown on adjoining premises. 



An interesting, but lengthy experiment might be made 

 with canna seeds by packing them in sand, using some kind 

 of indestructible container such as a perforated concrete box 

 that would exclude underground trespassers but allow a free 

 circulation of moisture in the box. The seeds should be so 

 arranged that each would l)e separated from its neighbor bv 

 an intervening layer of sand. The contrivance could then 

 be buried and a part of the seeds tested for germination at 

 intervals of five or ten years. Such an experiment would 

 very likely show whether the peculiarities related above are 

 constant or just mere freaks. 



PLANT NAMES AND THEIR MEANINGS— XV 

 SCROPHULARIACEAE 



By WiivLARD N. Cluti;. 

 OO like the toad-flax in appearance as to be joint owner 

 ^^ of some of the common names in the "snapdragon" 

 (Aiifiirliiiniiii ma jus). The first svllable in the name is as- 

 sumed by many to refer to the way the corolla opens and shuts 

 or "snaps" when pressed in just the right way, but Britten and 

 Holland assert that the name is really siicb dragon which is 

 literally "dragon's snout". It will not do, however, to jump 

 to the hasty conclusion that the common name refers to the 

 flower. An examination of a ripe seed-pod will disclose a truly 

 dragonlike countenance with staring eyes, a broad mouth, and 

 a long snout. It is quite possible that this is the dragon's snout 



