36o TRANS. ST. LOUIS ACAD. SCIENCE. 



on the tube of the corolla. The awned calyx- lobes must in 

 a measure prevent the insect from perforating the tube of the 

 ■corolla farther down. The same is true of Lamiiim album and L. 

 •amplexicaule. 



The habit of perforating flowers is often very local. I have 

 alluded to the perforated Howers of Monarda Bradburiana in the 

 Botanic Garden. Sometime later I looked for perforated Howers 

 of the same species in several places some twenty miles from St. 

 Louis, but I never found the flowers perforated, so that insects 

 are sometimes very local in their habits, as Darwin and others 

 have shown. Focke (40) also alludes to this localization in some 

 hybrids between Nicotiana rustica and N. paniculata (the parents 

 of this hybrid were not touched), in one patch of which the 

 flowers were perforated, while in another removed a little dis- 

 tance none of the flowers were thus treated until somewhat later, 

 and these were undoubtedly made by the same bee, yet both 

 patches where equally conspicuous. What should cause this lo- 

 calization of habit .^ It must, T think, be due to the individual 

 experience of the insects, for those which had once perforated 

 flowers always did so in this case. 



Why should insects perforate flowers? Darwin (25) believes 

 that, as a general rule, flowers are only perforated when they 

 grow in large quantities close together ; for he found in a garden 

 where Stachi/s coccinea and Pentstemon argutus were growing in 

 large numbers every flower was perforated, but at some distance 

 from these was a small stock of Stachys coccinea the flowers of 

 which were much scratched, showing that they had been visited 

 by bees, although not a single flower was perforated. The same 

 thing w^as noticed on a small stock of Pentstemon, growing in 

 the same garden. The same fact holds true in Trifoltuin pratense 

 when growing in fields, and Phaseoivs multiflorus when grown in 

 large and conspicuous clusters in gardens. It is a well known 

 fact that alpine flowers grow in much larger masses than plants 

 of lower regions. Familiar examples of our flora are aflbrded 

 by Silene acaulis, Arenaria Groenlandica, Bryanthus taxifolius, 

 TrifoUum sp. Ledum, etc. Miiller (90 A) has shown of alpine and 

 subalpine plants that more flowers were visited in the Alps than 

 in the lowlands, and also that more species were perforated, as 



