Speight. — An Ancient Buried Forest near Riccarton . 363 



the whole depth of this pit river-gravels and occasional irregular sandy 

 layers form the material of the beds, and there is an absence of the stratifi- 

 cation which would be evident were the beds of marine origin. Further, 

 they appear beyond the limit of the area where there is the interstratifi- 

 cation of pervious and impervious beds necessary for artesian conditions at 

 shallow depths, although this undoubtedly occurs at lower levels in the 

 same locality, as is indicated by the records of the wells at Islington (8). 

 There are similar occurrences of timber in an excavation for gravel close 

 to the Heathcote River, in Spreydon, at the foot of the Port Hills, as well 

 as in gravel-pits near the Harewood Road. 



A more interesting pit, however, is found in the Riccarton district, at 

 Sockbum, close to the Paparua County Council's office. This is about a 

 mile and a half nearer Christchurch than the Hornby pit. It is about 63 ft. 

 above the sea, and has been excavated to a depth of about 12 ft., at which 

 level a bed of sandy clay forms an impervious layer on which water lies 

 except in summer and autumn. The pit has an area of over an acre, and 

 all over this the stumps of trees, mostly of totara, are in position. (See 

 Plate XXIII, fig. 1.) Fully thirty large trees are thus represented, some up 

 to 5 ft. in diameter. At my first visit a large tree, 40 ft. in length and 

 over 4 ft. in diameter, lav exposed on the floor of the excavation, having 

 apparently broken off from one of the stumps : this was subsequently cut 

 up for timber. Dining a recent visit to the spot in company with Dr. 

 Cockayne I noticed another large tree, fully 5 ft. in thickness, which was 

 uncovered for 20 ft. or more. There is no doubt that we have exposed 

 here an old forest which was growing on the floor of the pit when that 

 formed the land surface, and which has been subsequently buried by the 

 pouring-in of sediment from adjacent rivers. It is clear that the standing 

 trees were broken off from their stumps before they were buried up, other- 

 wise the trunks would not be separated from their roots as they now are. 

 What caused the breaking-off is not apparent, but in all probability the 

 trees were killed while in the standing position and then the stems were 

 snapped off near the ground, or perhaps while the roots were covered up 

 with river-gravels. This process can be seen on many of the shingle fans 

 and river-beds in the mountain region of the province. Living trees of 

 mountain-beech or wild-irishman can be observed standing partly buried in 

 a waste of moving shingle, maintaining a precarious existence, while along- 

 side stand dead and partially buried standing trunks, and farther down- 

 stream the fallen stems lie at all angles and are buried up completely or 

 exposed as the stream washes out a deep channel on the surface of the 

 fan. The occurrence in the gravel-pit at Sockburn is exactly analogous to 

 the second and third of these conditions. 



No doubt the growth and destruction of the forests on the plains went 

 on simultaneously in adjacent areas and went on alternately in the same 

 area, the destroying gravel forming the substratum on which the next 

 forest growth was based. Stretches of land with good soil would occur in 

 certain parts, usually low-lying alluvial flats, and on these the forests would 

 establish themselves and maintain their footing for a time ; but these parts 

 would be especially susceptible to inroads of gravel, since ultimately the 

 river would use them for a dumping-ground, and the forest would be 

 destroyed or be compelled to remove to an adjacent area. The fact that 

 the forest must have been established on a land surface definitely supports 

 Haast's original idea of the formation of the plains, and of itself would 

 negative the idea that they had been formed of waste carried to sea by 



