480 NATURAL HISTORY OF 



Dahlhornii), already noticed, is common in these dreary 

 regions, and, owing to its size and the narrowness of the 

 aperture of the tube of the flower of Desfontainea, which is 

 blocked up by the anthers of the stamens, is unable to enter 

 by the mouth to suck the honey-like fluid at the base. It 

 therefore bites a hole in the side of the flower near the base, 

 and inserts its head there. We watched the process carefully 

 several times, as I was anxious to make out whether this 

 insect played any part in the fertilising of the flower, and I 

 finally came to the conclusion that it did not, as the stamens 

 are quite removed from the part of the corolla attacked by it. 

 On the 3d, leaving a party to finish the survey of Mayne 

 Harbour, we moved northwards to Puerto Bueno to execute 

 an uncompleted piece of work. Arriving there about noon, 

 we parted company with the " Eingdove," which continued 

 her northerly course. I passed the afternoon on shore, but, 

 with the exception of the discovery of a single dead and 

 worn Concholejpas shell, met with nothing of interest. We 

 returned to Mayne Harbour next morning, and remained 

 there for the rest of the day. We weighed on the morning 

 of the 5th (one of the most utterly dismal days we had yet 

 seen, with the hills shrouded in mist, heavy rain descending, 

 and occasional tremendous squalls), and, moving southwards, 

 anchored in Columbine Cove, in Piazzi Island, in the after- 

 noon. The following morning was tolerably fine, but a good 

 deal of snow had fallen during the night, and the tempera- 

 ture was decidedly frigid. I spent an hour on shore, while 

 some of the officers were taking sights, and walked over the 

 narrow peninsula which separates the cove from Shingle 

 Koad, but observed nothing remarkable. The only sign of 

 life was furnished by the little Cinclodes Patagonicus, which 

 was diligently searching the masses of kelp on the beach for 



