320 THE JOURXAL OF BOTANY 



the hive-bees did not visit the flowers of the Primrose. On the 

 present occasion Apis mellifica was collecting from various flowers, 

 and I observed it at work on White Rock (Arabis alpina), Wall- 

 flower, American Currant (Ribes sanguineum), Dandelion, Barren 

 Strawberry, Gorse, and Willow. No humble-bees came to the 

 Primroses, though they frequented the Willow catkins and other 

 flowers. Although the grassy bank upon which a special watch was 

 kept was occupied by large and conspicuous groups of Primroses, the 

 only visitors observed during the two hours were numerous sin ill 

 Coleoptera (Meligethes picipes) and a fly (Lucilia ccesar). The 

 latter was only noticed as an occasional visitor, and appeared to be 

 resting rather than attempting to obtain food. I observed that this 

 Dipteron and certain other flies constantly visited numerous inter- 

 spersed Daisies for pollen or nectar, or both. The following day 

 was fine and sunny, but a chilly wind prevailed. On this occa- 

 sion the grassy banks of the lane to Khuallt were kept under obser- 

 vation for some time (11 to 1.0). Thousands of flowers occurred 

 along the half-mile or so distance regularly patrolled, but I did 

 not observe a single bee, nor were other winged insects seen on the 

 Primroses. The flowers here, as at Bryn Goleu, were tenanted by 

 numbers of a species of Thrips ; this appears to occur wherever the 

 Primrose is met with, and is doubtless a factor, as Darwin admits, in 

 assisting self-pollination of both types of flower. 



It seems rather remarkable, especially in view of Darwin's state- 

 ment that the Primrose is pollinated by moths, that no careful 

 actual observations have been made at night-time. As the plant 

 shows certain features of Midler's " lepidopterid flowers," it seemed 

 desirable that investigation should be made. My interest had been 

 further aroused by noticing numerous instances of Primroses in which 

 the corolla had been more or less gnawed or devoured ; in some cases the 

 flower had been attacked to such an extent that only the corolla-tube 

 remained. In many cases (especially in the short-styled type) stamens 

 and occasionally the entire androecium had been removed ; in other 

 examples (particularly in the long-styled type) the stigmas had dis- 

 appeared. Such mutilation is not peculiar to the Cwm area; I have 

 subsequently observed it in many localities in counties as far apart as 

 Denbigh, Bute (Isle of Cumbrae), Dumbarton, York, Ayr, Cheshire, 

 and Renfrew. Some of the mutilated flowers about Cwm (as others 

 elsewhere) bore evidences of the visits of slugs or snails, and empha- 

 sized the desirability of nocturnal observations. A row of fifteen 

 plants, with perhaps about 110 open flowers, on a grassy bank at 

 Bryn Goleu, was visited on three nights (April 14, 15, 17) and 

 patrolled at intervals for periods varying from fifteen minutes to an 

 hour between 8.30 and 10.45 p.m. An electric flash-lamp was used 

 to inspect the flowers periodically. Taking these three occasions 

 together, the flowers were found to be visited by the following : — 

 6 Forficula auricularia N., 3 Meligethes pieipes Sturm, 2 Otiorhgn- 

 chus picipes Fab., 2 Limax agrestis L., 1 Otiorhynchus ligneus Oliv., 

 1 Sciaphilus muricatus Fab. 



The three n'ghts were rather chilly ; on the 14th and 15th 

 there was a cold wind which was very unfavourable for inv purpose. 



