Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 26, 1893
This text is also available in PDF
(753 KB) Opens in new window
– 226 –

Art. XX.—Result of a Further Exploration of the Bonefissure
at the Castle Rocks, Southland.

[Read before the Otago Institute, 14th November, 1893.]

Plates XXIII., XXIV.

In the Transactions of last year I gave an account of the Castle Rock fissure, and of the specimens yielded by an exploration of its depths.

This year Mr. Mitchell and myself have, through the kindness of Mr. Barnhill, renewed our work, with very satisfactory results, the anticipations expressed in the previous paper having been fully realised.

It will first be necessary to take this opportunity of making a few corrections and additions to the diagrams as printed in the Transactions, vol. xxv., pl. vii. The letter C (see p. 90) should be placed more to the left of the person looking at the diagram, on the dotted portion. In the diagram it is on the limestone rock. The letters D, E, and F have been omitted altogether. They should occur on the plan, and on the section from 3 to 4, D being just under the figure 4, E at the narrow gap at the other end, and F marking the arm of the cave near the figure 4. A more important error runs through the tables of dimensions (with the exception of that of Ocydromus), “Specimen No.” being printed for “Number of Specimens.”

Considerable preparations were made by us for this trip, as I was fully convinced that the majority of the missing bones of Harpagornis were there if we could only find them. Our first efforts were directed to trying to devise some method of taking the soil out of the fissure altogether and spreading it on the surface at the entrance. This necessitated the clearing-away of a number of small bushes and “lawyers” to give an open space on which to spread and examine the soil. This being done, ropes were rigged, and two iron buckets. When filled these were drawn up to the surface, emptied, and raked over, so that any small bones contained might be seen. This gave us a considerable number of small bones which would not otherwise have been obtained, but it was slow work, and was only carried on until sufficient of the loose and previously-worked soil had been brought up to allow of further examination of the “lead” of bones under the rock. When we left off last time the material excavated from the trench kept slipping down again into the trench, and was a source of danger. Now, however, it was possible to go much deeper, and for 8ft. or 9ft. down we obtained bird-bones, principally

– 227 –

of the duck and kiwi. At the very lowest depth reached there were still a few of these to be obtained, but there seemed no chance of reaching the actual bottom, so we ceased descending, and went into the small cave through the narrow passage. The floor of this was dug over again, and a number of kiwi-, duck-, and kakapo-bones obtained.

At p. 90 (loc. cit.) of the Transactions I mentioned that the lateral arm of the cave (which should have been marked F on the plan) contained a quantity of a wet, soapy deposit consisting of nearly pure carbonate of lime. This was examined to a much greater depth than on the previous occasion, and at least 2ft. of the deposit of carbonate of lime removed from the surface, which did not contain a bone. Beneath this we came on a layer of clay and lime containing bones, and this led us to examine it very carefully, as one of the first bones of any consequence proved to be of Harpagornis.

I had the whole of the lime removed first, and then carefully worked out the original floor. The lime was in some places quite hard, like well-set plaster of Paris—in other places quite sodden and soapy. Starting from the corner of the main fissure, bones were found to occur plentifully, and to extend downwards for about 2ft., or about 5ft. below the original surface. The first find of importance was a capital skeleton of an individual moa—a young individual—with skull, pelvis, legs, &c., close together.

Whilst I was numbering and marking these in the light word was passed out from the workers that eagle-bones were being found. Extra candles were lighted, as this arm of the cave was quite dark. I soon found that several important bones were exposed, their smooth yellow surface contrasting with the pure white of the lime, which at this place formed a small “pocket” in the clay floor close to the rock. With a pocket-knife I removed the dirt, and had the pleasure of finding the greater part of the skeleton of the smaller of the two eagles usually known as H. assimilis. The skull (Pl. XXIII., fig. 1) was remarkably perfect, lying on its vertex, with the lower mandible in position, and one quadrate. Across it lay a humerus, which had protected the delicate bones of the skull from injury. The legs and both wings were found here. A little further on the sternum was found quite perfect, but slightly cracked (Pl. XXIII., figs. 2 and 3). It was curious to note that in the cranium of the skull there still remained several of the chitinous pupa-cases of the flesh-flies that consumed the fleshy remains of the lordly eagle. A few separated bones of the larger bird (H. moorei) were then found; and just as the work of cutting into the hard deposit was being abandoned the workers found a small hole just large enough to admit a hand. This led into a small extension of the

– 228 –

fissure, and several other bones were obtained. At the very end, firmly fixed in the hard carbonate of lime, the edge of a large bone appeared, and I recognised this as the sternum of H. moorei (Pl. XXIII., fig. 4); so we spent a considerable time in cutting out with an axe sufficient of the sides of the fissure to enable a block to be detached at the end containing the sternum. The block was then carefully removed and packed away. Work at the main fissure was renewed next day, but without much success. Another almost entire skeleton of a young moa (Anomalopteryx) was obtained in good preservation, and, as usual, a vast number of small bones of duck, kiwi, kakapo, &c.

The most curious and interesting specimen found here is the sternum, figured at Pl. XXIV., fig. 1, of Anomalopteryx-didiformis.

Amongst the remains described in the previous paper was a very good skeleton of an adult A. didiformis, which is at present deposited in the Otago University Museum. It is remarkable in having a process on the right side of the upper edge of the sternum of an irregular form, about 25mm. in height (Pl. XXIV., fig. 3). I concluded at the time that it was an occasional or abnormal ossification; but now another specimen has turned up in which the process is present on both sides, about 40mm. in height, and corresponding in shape with each other. Having now re-examined the specimen in the Museum, I think there is reasonable ground for supposing that it also possessed two of these processes, the place on the left side corresponding to the base of the process having a surface indicating an irregular articulation, as in Aptornis—a non-synovial union.

I have since examined several other sterna of the same species, and many of them appear to have originally possessed something of the same kind of episternal process.

It is quite uncertain what form the scapulo-coracoid took in these small moas, as in the whole series of bones from this fissure I have not found a single bone that can be said with certainty to be a scapulo-coracoid or to belong to the more or less defined pits on the upper edge of the sterna which I have considered coracoid notches. I have carefully examined a number of specimens of the same species from Enfield, and, although I have not seen any with the process remaining, there were several that may possibly have had it, that portion of the sternum being abraded or damaged so that no definite opinion could be formed.

Notwithstanding the apparent coracoid notches at the back of the process, I think these irregular prominences must be anchylosed scapulo-coracoids, and that the depressions, which vary greatly in depth and size, are merely the result of the

Picture icon


Picture icon


– 229 –

development of the adjacent part. It is to be noticed that the foramina at the back of the sternum in this region are excessively variable in size, being larger and more frequent as the indications of a coracoid are greater in the front of them. In the specimen figured (Pl. XXIV., fig. 1) there is quite a large pit or lacuna in the rear of the depression behind the processes. The hinder border of the middle portion of the sternum (No. 256, fig. 3) is deeply notched, as in some sterna of Apteryx.

Behind the costal processes on the neural surface of this sternum there are two small processes (a, fig. 3) directed inwards. The same process occurs in No. 313, but only on the left side.

Amongst the small bones I have two fragments resembling the processes, one of which may have belonged to the sternum of the specimen in the Museum. In my next paper I hope to give a résumé of the whole collection, and to include the references to Harpagornis in Dr. Von Haast's third paper, in vol. xiii. of the Transactions, which I seem to have over-looked.

Description Of Plates XXIII., XXIV.
Plate XXIII.
  • Fig. 1. Skull of male Harpagornis; extreme length, 150mm.

  • Fig. 2. Side-view of sternum of male Harpagornis.

  • Fig. 3. Front-view of sternum. Length, 145mm. = 5·7in.; width, 87mm.

  • Fig. 4. Lower margin of sternum of female Harpagornis. Length, 165mm.=6·5in.; width, 90mm.

Plate XXIV.
  • Fig. 1. Sternum of Anomalopteryx didiformis, No. 313, from Castle Rock, with anchylosed scapulo-coracoids (?).

  • Fig. 2. Side-view of left scapulo-coracoid.

  • Fig. 3. View from the back of the anchylosed scapulo-coracoid on the right side of the specimen No. 256, deposited in the Otago University Museum; a, small process from costal.