KILAUEA. 193 



lhese roots not only grow from the side which requires support, but 

 seem to take that direction which will likewise furnish soil. When the 

 pandanus forests are in full bloom, the whole air for miles around is 

 scented with the fragrance. 



This day, for the first time, I saw a deranged native. He had 

 escaped from his keepers ; and I thought he was rather harshly used 

 in what I saw of the affray. There was gi'eat difficulty in securing 

 mm. 



Within a mile or two of the observatory, we met Mr. Drayton, going 

 out on another excursion. 



We reached the observatory after an absence of forty-two days, 

 and it was delightful to feel ourselves as it were at home again, after 

 so arduous and fatiguing an expedition. I had the pleasure to hear 

 that every thing had arrived safely, and that all were well. On 

 inquiry being made for the Kanaka that had been missing, I learned 

 that a party of natives had gone to the mountain in search of him, but 

 that little hope was entertained that he would be found. 



On the 24th, Lieutenant Carr, Dr. Fox, Lieutenant Case, and several 

 of the other officers, were allowed permission to visit the crater. 



On the 25th, 26th, and 27th, we were employed in putting up the 

 pendulum apparatus, and began on the night of the latter to observe 

 coincidences. The three series obtained, to my great astonishment 

 did not agree. I could not account for the discrepancies, for I exa- 

 mined the whole apparatus, agate planes, knife edge, &c, and found 

 them in perfect adjustment, both with plumb-line and level ; the scale 

 and telescope were also found correct ; the rate of the clock was 

 steady under hourly comparisons with the siderial clock, and observa- 

 tions of the transits of stars. 



Although I had covered the pendulum-house with its tent, I thought 

 that some of the discrepancies occurring might be owing to its want 

 of uniformity of temperature. I therefore, on the 29th, had a grass- 

 house built over both, which remedied any defect of temperature. I 

 then continued to observe, but did not find the results more satis- 

 factory ; I therefore took down the whole apparatus, put it up, read- 

 justed it anew, and took another series during the night. These were 

 rather more in accordance. I continued observing through the 30th 

 and 31st, but with no satisfactory results. I then examined the pen- 

 dulum again with the plumb-line on both sides at the same time : they 

 both coincided with the marks made in London by Mr. Baily and my- 

 self, in 1836. I next tried the iron pendulum, and found it to agree 

 also ; reversed both pendulums on the knife-edge, but the results were 

 still discrepant. 



vol iv. k 25 



