May 6, 1880] 



NATURE 



13 



due in a great measure to his advice. Need we wonder, 

 then, even with vouth in his favour, that at the early age 

 of about thirty-seven, being yoked in such heavy double 

 harness, he has died with it on, leaving a large amount of 

 accomplished valuable work, which was waiting for a 

 spare moment to prepare for the publisher. 



For some time Dr. Scheffer had been suffering from 

 defective digestive powers and frequent sleeplessness, 

 but he neglected these warnings and the advice of his 

 friends to take some rest. He was unfortunate in bemg 

 surrounded by those who, with few exceptions, took little 

 interest in his work, and by none to whom he thought he 

 could entrust the work in which he was so hard and 

 enthusiastic a worker, so he worked on. The fatal afiec- 

 tion was inflammation of the liver. The seizure was very 

 acute, and at an early stage danger was immmcnt ; but 

 at length he rallied. His medical attendants considered 

 the crisis past, and recommended his removal to his own 

 estate near Sindanglaya, to reach which a tedious climb of 

 4,500 feet over the Megameudoeng Pass had to be sur- 

 mounted. He never reached his destination, expiring, on 

 March 9, at the Sanatorium at Sindanglaya, where he 

 now lies buried. 



In his private life he was a man to be loved and 

 esteemed ; quiet, unassuming, very kind-hearted, ever 

 ready to give whatever assistance he could, especially to 

 scientiflc travellers. With him the Netherlands Indian 

 Government has lost a valued public servant, to whom it 

 will not be easy to find a successor, and botanical science 

 has to deplore an earnest worker, a learned disciple, and 

 a great helper. Henry O. Forbes 



Preanger, Java 



V 



.4 SCOTTISH CRANNOG^ 



V 



BETWEEN geology and history there lies an inter- 

 mediate sphere in which these sciences dovetail into 

 one another. In this common territory or borderland 

 lies the domain of prehistoric archeology, and to its most 

 recent portion, or that which archaeologists have desig- 

 nated the " Late Celtic Period," must be assigned the 

 antiquarian remains I have here the pleasure of describing. 

 During this period it appears that the Celtic races of 

 Scotland and Ireland were in the habit of constructing 

 artificial islands in marshes and shallow lakes to which, 

 in troublous times, they resorted for safety. They were 

 generally formed by the superposition of trunks of 

 trees and brushwood mingled with stones strongly 

 palisaded by stakes, and so situated as to be inaccessible 

 e.xcept by means of causeways, or occasionally by a narrow 

 gangway or mole. These island forts, or craunogs, as 

 they have been called in the Irish annals, were very 

 numerous in former times, but owing to the gradual rising 

 of the level of the lakes, they appear to have been so 

 completely lost sight of that their very existence was 

 unknown to modern antiquaries, so that their discovery 

 in the present century marks an important epoch in the 

 history of archaeology. 



In October, 1S78, I drew the attention of antiquaries, 

 through the columns of Nature, to the remains of ;^an 

 ancient lake-dwelling just then discovered on the farm of 

 Lochlee, in the parish of Torbolton, Ayrshire. Since then 

 a series of excavations have been made with the view of 

 ascertaining the exact nature of this structure, in the 

 course of which a large collection of most interesting 

 relics has been made. 



In the year 1839, while a small lake on this farm was 

 being artificially dried up for agricultural purposes, the 

 attention of the labourers was directed to a singular 

 mound, in which, on cutting drains through it, they 

 exposed some wrought wood- work; but their observations, 



' A full report of the Lochlee Crannog is given in vol. xiii. of the 

 rrocceiiiiigs of the Society of Antiquaries of Scotland, and in vol. ii. of the 

 Collections cf the Ayrshire and Wigtownshire Archaeological Association. 



though freely talked of in the neighbourhood at the time, 

 led to no further results till forty years later, when it was 

 found necessary to re-drain the locality, and hence the 

 present investigations. By a curious coincidence the 

 early drainage at Lochlee was made in the same year 

 that Sir W. R. Wilde discovered and examined the first 

 Irish crannog, viz., that of Lagore in County Meath. The 

 Irish discovery, however, owing to a general system of 

 drainage that was then going on, led at once to the most 

 brilliant results, so that it soon became apparent that 

 crannogs existed very generally over the country. Up to 

 the present time over a hundred have been examined, and 

 have furnished the Irish museums with a vast collection 

 of relics. In the year 1854 a great impetus was given to 

 the study of these researches by the discovery of the 

 remains of ancient lake villages in Switzerland, which 

 have now become so famous and well known all over the 

 continent of Europe; but it was not till 1857 that the 

 subject began to attract the attention of Scottish archaeo- 

 logists. In this year Mr. Joseph Robertson read a paper 

 to the Society of Antiquaries of Scotland, and in i866 

 Dr. Stuart, who was then Secretary to this Society, 

 collected and published all the scattered notices of Scottish 

 crannogs known up to that date. Since the publication of 

 Dr. Stuart's elaborate paper no further investigations on 

 Scottish crannogs, with the exception of an occasional 

 notice of a fresh discovery of Jhe site of one, have been 

 recorded. 



But though traces of these crannogs, have been found 

 in almost every county of Scotland, there has been no 

 svstematic examination of them worthy of comparison 

 with the investigations that have been made in other 

 countries ; nor, with the exception of a few articles found 

 at Dowalton, is there any collection of relics which would 

 enable archaeologists to form an opinion with much 

 certainty as to the purpose they served in the social 

 economy of the period they represent; nor can their 

 range in the dim vista of prehistoric times be determined 

 with greater accuracy. 



Before the Lochlee Lake was originally drained no one 

 appears to have surmised that a small island (visible 

 only in the summer time) which formed a safe habita- 

 tion for gulls and other sea-birds during the breed- 

 ing season, was formerly the residence of man. It was 

 situated near the outlet of the lake-basin, and the nearest 

 land, its southern bank, was about seventy-five yards 

 distant. The general appearance which it presented 

 when the present investigations were commenced was 

 that of a grassy knoll, 'drier, firmer, and slightly more 

 elevated than the surrounding field. Towards the cir- 

 cumference of this mound the tops of a few piles were 

 observed barely projecting above the grass. Guided by 

 these the workmen dug a deep circular trench, m which 

 they exposed numerous piles and transverse beams having 

 square-cut holes in their ends, through which the former 

 projected about eighteen inches or two feet. In the 

 course of further explorations it became apparent that 

 these piles formed a series of stockades surrounding a 

 somewhat circular space about fifty feet in diameter. 

 Beyond this circle on the south side there were indications 

 of other rows of uprights which appeared to unite into 

 one on the north side. Here, instead of further rows of 

 piles, the corresponding space was occupied by an 

 intricate arrangement of woodwork, consisting of young 

 trees and stout branches, mixed with slanting stakes and 

 logs running in all directions, the whole forming a dense 

 pi-?)tective barrier. The diameter of the island was about 

 1^0 feet. The central area was about three feet lower 

 than the surrounding stockades with their transverses, 

 and had a flooring of prepared logs resembling railway 

 sleepers. Near the centre of this log pavement were found 

 four circular hearths placed one above the other with an 

 interval between each of 18 inches to 2i feet. These 

 hearths were neatly constructed of flat stones of various 



