(4.) That they were formed in situ by deposition of P₂O₅ from descending waters which derived their P₂O₅ from the basalt.

(5.) That they were formed in situ by deposition of P₂O₅ from percolating waters which derived their P₂O₅ the limestone.

(6.) That they were formed in situ by the concentration of the phosphatic contents of the limestone—a process due to ordinary weathering and the action of meteoric waters.

(7.) That they were formed by the combined action of the methods outlined in the last two theories. This combination theory is the one I see most reason to support.

1. Transportation Theory. — This is extremely improbable. The phosphate boulders are separated from the limestone and from each other by bands of phosphatic clay and by layers of sand. Running water would most probably have been the agency by which the boulders were transported, if they were transported at all. It is scarcely possible that a violent stream capable of moving a boulder of 2 or 3 tons weight would deposit that boulder on a thin stratum of clay or sand. Much more likely is it that the stream would sweep away clay and sand and deposit the boulders on the bare surface of the limestone. I have not observed a single instance of this direct contact of limestone and rock-phosphate; I have always found a layer of clay or sand between the two. Moreover, the presence of basalt-fragments imbedded in the phosphate at Kiln Point is sufficient to disprove the theory, for it shows that the rock-phosphate was formed after, and probably a long time after, the extrusion of the basalt.

2. Concretionary Theory. —It is not at all probable that the rock-phosphate was formed by concretionary action around some nucleus, in a manner analogous to the coprolites of the Upper and Lower Greendands of England. None of the Clarendon nodules have been found to contain organic nuclei, and the sharks' teeth and other organic fragments at Round Hill, &c., always lie on the clays, and not imbedded in the hard nodules.

3. Asension Theory. — The occurrence of phosphorite and perhaps staffelite as stalactites and as incrustations on the walls of cavities suggests that an aqueous solution of phosphoric acid must have played, and must still be playing, some part in the formation of the phosphate. Could this solution have risen from below? No, it could not, for in many places the phosphate rests on the surface of unaltered limestone; had the solution risen from below it would first have acted on the lower layers of limestone, converting the calcium-carbonate to

calcium-phospahate, and the process would then have extended itself upwards. As we find the phosphate above and the carbonate below, it is tolerably certain that the process is extending itself downward, and is due to descending solutions.

4. Basalt Theory. —That the deposits were formed in situ by the deposition of P₂O₅ from descending waters which derived their P₂O₅ from the basalt. I do not think it possible that the overlying basalt could have supplied the P₂O₅ necessary. No phosphorus-bearing mineral was detected in the microscopic examination, and micro-chemical tests likewise failed to reveal the presence of any such mineral.

5. Limeston Theory. —That the deposits were formed by the deposition P₂O₅ from percolating waters which derived their P₂O₅ from the limestone. It is certain that the phosphoric acid was derived from the limestone. We have seen that organic remains rich in phosphorus are found in the latter, and that, moreover, a small amount of phosphate is distributed through out its mass. I think that a deposition from waters is in part responsible for the formation of the phosphate, but that it is subordinate and subsequent to a concentrating action.

6. Concentration Theory —That the deposits were formed by the concentration of the phosphatic contents of the limestone, by the weathering action of waters containing carbonic and perhaps other organic acids, which dissolved out the calcium-carbonate of the limestone, but left behind the much less soluble calcium-phosphate.

7. Combination Theory. —That the process just outlined was followed by the deposition of P₂O₅ from percolating waters which had leached out their P₂O₅ from the limestone. The mode of weathering of the limestone at the Millburn quarry lends a great amount of support to this view. The surface of the limestone is carved out into a number of deep “guts,” leaving lofty pinnacles and overhanging shelves of limestone—an appearance which at first suggest a striking unconformity with the brown sands which fill the depressions. The sculpturing, however, is almost wholly if not entirely due to chemical erosion.

On the outer parts of the pinnacles, where exposed to wind and rain, the limestone has in many places weathered to crumbly brownish sandstone, containing comparatively a small percentage of CaCO₃. Along the laminæ the weathering has progressed more quickly. Isolated “floaters” of limestone are found in the middle of the brown sands; they are blocks of limestone which have offered great resistance to the weathering process, and are now entirely surrounded by clays and sands, in the midst of which, still horizontal, they seem to float. A sample was taken from the interior of one of theses floaters, where it was least

Weathered; it was partially analysed, as also was a sample of absolutely fresh limestone of the same band (H). The analyses gave:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

These analyses prove, (1) that the limestone H contains when fresh a small amount of phosphoric acid disseminated through its mass; (2) that the ordinary process of weathering, which carries away the calcium-carbonate in solution, does not carry away the calcium-phosphate to an equal extent, so that the latter tends to concentrate in the residue.

The “Gut at Millburn. —Brown and green sands and clays fill the hollow between two pinnacles, and protect the limestone surface from the atmosphere, but not from percolating waters. In the deepest central “gut” the limestone is found to have a thin veneer, about 0·75 in. thick, of a pulverulent weathered limestone (C₁). Outside this veneer comes 6 in. of a green glauconitic and phosphatic sand (C₂), the whole of this layer being laminated with fine streaks of not yet completely weathered limestone, arranged rudely parallel to the present limestone surface. Outside C₂ is found a varying thickness of yellow-green very sandy clays (C₃), laminated parallel to the surface of the limestone, the laminæ being from 0·06 in. to 0·25 in. apart, and composed of alternating green and yellow bands. These clays line the limestone wherever the latter is protected from the atmosphere —- on pinnacles, in hollows, and on the under-side of projecting ledges alike. A tough yellow richly phosphatic clay (C₄) is found in the midst of the yellow-green clays; it is sometimes in bands, more often in irregular patches and nodules.

The following analyses will show the manner in which the concentrating action has taken place. The limestone has been deprived of its CaCO₃; the quartz, glauconite, and to a great extent the lime-phosphate, being insoluble, have been left behind and concentrated to form C₁ and then C₂. Then the C₂ has been separated into two parts, the lime-phosphate tending to segregate into the clayey nodules C₄, surrounded by yellow-green clayey sand C₃. This differential action is due to water, which has acted by dissolving and then reprecipitating some of the lime-phosphate.

The above analyses show that a concentration has taken place. If we change their form slightly they show us also that the calcium-phosphate is not absolutely insoluble, and that even in the process of concentration some of it has been carried away in solution. In the concentration and solution it is safe to assume that the silica (SiO₂) is not carried away nor affected by any chemical action. We are thus enabled to recalculate the analyses on the basis of constant SiO₂: e.g., if a body (x) contains 1 part per 100 of an unchangeable substance (m), and a quantity of it is partly dissolved away, resulting in a product (y) which contains 5 parts per 100 m, then it is clear that the product y results from 5 times its weight of the original x. On recalculating thus from the analyses we find that 186 parts, C₁ have weathered to 29 parts of C₂, which have differentiated into 14 parts C₃ and 15 parts C₄. The following table gives the amounts of the constituents of each of these products:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

The C₁ column does not tally exactly with the results from the other columns, and this is probably due to errors in the chemical analysis. It must be remembered also that any errors that do occur in the analysis are almost doubled in the case of C₂, while they are diminished in the case of the others, roughly in the proportion of 3 to 1 and 6 to 1 respectively. Noticing the other columns only, we see that the iron and alumina, like the silica, are not removed by the weathering process, 5·35 resulting from 5·36. The amount of moisture (H₂O) is of no importance in this inquiry, though apparently it also does not change much. P₂O₅, however, has been abstracted, 1·33 resulting from 1·72. This difference is too great to be merely due to error in analysis. It is clear, then, that phosphoric acid has been dissolved out from the clays; it joins the downward circulation, and is redeposited where conditions are favourabel.

Stages of Process.— The amount of action that has taken place in this gut at Millburn is small, but the full sequence of operations is clearly seen. At Wilson's quarry the process is more advanced, and has resulted in the formation of a large quantity of phosphatic clays, which are now being cemented together by the redeposition of the phosphoric acid as lime-phosphate in the cracks which traverse the clay. The process has reached its final stage at the Round Hill quarry, where all the calcium-carbonate of the limestone has been dissolved away, its place being now occupied by a mass of rock-phosphate. The final stage has likewise been reached on the right-hand side of the limestone quarry at Millburn, where now two boulders of hard rock-phosphate, in the middle of the brown sands, rest on the chemically eroded surface of the limestone.

Objections. — The objections which may be advanced against this theory are, I think, the following:—

1. That the process outlined cannot account for large quantities of rock-phosphate such as are found at Round Hill. To explain such a large occurrence it is only necessary to suppose that the original limestone at that point contained a very great number of organic remains, such as bones, &c., and this supposition is upheld by the occurrence of numerous bone-fragments among the phosphate at Round Hill.

2.That the rock-phosphate ought to be always accompanied by brown sands or sandstone. As a matter of fact it usually is, though it is no necessary that it should be so; the original limestone now eroded away may at one place have been rich in phosphate, poor in silica: this would produce a rock-phosphate with very little sandstone about. Conversely, the original limestone at another place may have been rich in silica, poor in or destitute

of phosphate: this would produce a sandstone with no rock-phosphate in its neighbourhood.

Conclusions.—My conclusion, then, is that the rock-phosphate has been formed by the action of meteoric waters slowly weathering away a limestone containing a small amount of lime-phosphate, leaving most of the latter behind, but dissolving some and reprecipitating it where conditions were favourable. Where the limestone originally contained a large amount of lime-phosphate, owing to abundance of vertebrate remains, there the rock-phosphate will be found in greatest abundance. Where the limestone originally contained practically no phosphate, there its weathering will give rise to no deposits of rock-phosphate. The limestone owes its contained lime-phosphate to the presence of organic remains, especially those of vertebrate animals; the distribution of these is of an irregular nature; the weathering of such a limestone ought to give rise to irregularly distributed deposits of rock-phosphate: this is what we find.

Professor Park,*^* in his account of the Clarendon phosphates, does not hazard any opinion as to how these particular deposits originated. He says, “The formation of phosphate-deposits is generally believed to have been due to the leaching or lixiviation of phosphate-bearing rocks by waters containing carbonic and other organic acids, followed by the subsequent concentration of the phosphate under favourable conditions. In Some cases they deposited their calcium-phosphate in caverns formed in lime stone or calcareous sandstone, and the subsequent removal by solution of the walls of the caverns, either wholly or partially, left the phosphate in the remaining sands.”

Similar Theories Abroad.—The theory advanced above resembles the theories which have been advanced to explain the origin of other deposits of lime-phosphate. Accounting for the phosphatic beds near Mons, Belgium, F. L. Cornet^† says, “These phosphates have been formed from the concentration of phosphatic matter, originally disseminated in the lime-carbonate, the concentration having been effected by the action of water containing carbonic acid.”

C. W. Hayes,^‡ accounting for the white phosphate of Tennessee, sums up thus: “The original lime-phosphate. accumalated with other sediments, either segregated in beds or disseminated through limestones and shales… These were attacked by percolating surface waters which contained carbonic and other organic acids, and which dissolved the C_aCO₃,

and in less quantity the Ca₃(PO₄)₂. The carbonate was carried away; the phosphate was redeposited, the form of the deposits being modified by local conditions.”

Comparison with Other Deposits.—The phosphates of South Carolina^* consist of waterworn nodules, much bored by marine animals; phosphatic casts of the interior of shells are abundant. The North Carolina Phosphates^† are in part like the South Carolina deposits, but there is also a phosphatic conglomerate with teeth, bones, nodules, and quartz pebbles all well rolled and rounded and cemented together. The Alabama deposits^‡ are, like those of South Carolina, composed of shells, phosphatic nodules, shell-casts, and fossiles, all much worn and broken, usually flat in form, and more phosphatized on one side than the other. the nodular deposits of Florida^§ rest on the uneven surface of a calcareous rock, being associated with shells and sands. At one of the quarries the soft calcareous rock gradually blends at a depth of 3ft. or 4ft. into a massive compact phosphate rock, similar in appearance to the phosphatic fragments above, except that it is a solid mass; it is probably the ledge whence the fragments were derived, the phosphatic pebbles and the sand being due to deposition on the eroded surface of the calcareous rock. Carnot,^∥ however, holds a different opinion of these Florida deposits. He considers them of concretionary origin; he emphasizes the high percentage of CaF₂, and argues that the concretionary action is due to the concentrating action of salt water. The phosphates of the West Indies, at Aruba and Sombrero,^¶ and also those of Ocean and Pleasant Islands,^** were originally coral limestones, now converted into phosphate by the percolation of waters containing phosphoric acid derived from the overlying deposits of birdguano. The deposits of Ottawa (Canada),^†† and of many other localities, occur as veins of apatite in very old igneous rocks, mostly of Archæan age. The phosphates of Wales^‡‡ consist of apatite veins and of amorphous nodular deposits; the latter contain numerous remains of animal life, and are due to the phosphatization of a calcareous bed. In England^§§ the de-

posits consist almost entirely of nodules which are composed of phosphatized animal matter, while those in the neighbourhood of Taplow and Lewes^* are phosphatic chalks of sedimentary origin. Those of the Somme in France^† are also chalk-deposits, the phosphate being “a sedimentary chalk derived from the disintegration of a vast granitic phosphatic continent.” Near Mons, in Belgium,^‡ the phosphate-deposits are also with Cretaceous chalk. They consist of a coarse-grained rock formed of a mixture of grains of calcite and small-sized pebbles of phosphate; they are due to a concentration by water of a limestone originally containing a small amount of lime-phosphate. In Algeria and Tunis^§ the phosphates occur in nodules in marl or as phosphatic limestone.

Though not precisely similar to any other known deposit, the Clarendon-Millburn phosphate bears greatest resemblance to those of Florida, U.S.A. The local phosphate, however, contains only a trace of fluorine, and has no shell-traces; the calcareous rock does not become highly phosphatic in depth; and the mode of origin is different from either of the two theories advanced to explain the Florida deposits. In all these points it differs from the Florida phosphate, so that, after all, the similarity is not so very marked.

Mining and Treatment.—At Round Hill quarry the phosphate is blasted out with gelignite, the holes for the charges being very difficult to bore, on account of the hard nature of the rock. Very little hand-sorting takes place in the quarry, the rock being immediately loaded into trucks or “boxes” and drawn to the burning-ground, where it is built up with broadleaf timber into large piles. The burning extends throughout several days, and deprives the rock of its moisture, thus rendering it brittle so that it can be more easily pulverised, and also effecting a small saving in the cost of its railway carriage to Burnside, a few miles south of Dunedin, where some of it is treated with a spray of dilute sulphuric acid to partly convert it into superphosphate of lime. This chemically treated phosphate commands a higher price than the phosphate which has not been so treated. The former is more rapid in its effect on the crops, as the superphosphate is at once ready for assimilation by plants, while the phosphate in the crude, unsprayed rock has

to be slowly converted by the action of humic and other acids before it is available for absorption.

Prospects of the Industry.—The amount of rock-phosphate occurring in the district cannot be accurately estimated at present. As the original phosphate to which it owes its origin was probably irregular in its distribution, so we must expect our deposits of rock-phosphate to be irregularly distributed along the outcrop of limestone. The question whether the phosphate will occur between the already located outcrops cannot at present receive a satisfactory answer; in some places it will, in others it will not; the position of these respective places can only be determined by further prospecting. From its origin it follows that the phosphate will not extend inwards under the protecting basalt cap; and for this reason also it follows that there is a greater chance of its occurring in quantity under a gently sloping surface than under a steep one, for in the former case a much greater width of limestone has been exposed and subjected to the actions which lead to the formation of the phosphate. Until prospecting affords more information about the extension of the outcrops it would be useless to attempt an estimate of the quantity in sight.

The sale of the phosphate will, in my opinion, be confined to New Zealand. To compete in foreign markets, high-grade rock-phosphate must contain at least 77 per cent. of calcium-phosphate—Ca₃(PO₄)₂—and its amount of alumina and iron must be low. A reference to the list of analyses in this paper will show that it will be difficult to guarantee that any large quantity of the phosphate fulfils these requirements. Other countries, moreover, owing to their more favourable geographical position, are better enabled to command the foreign markets. In New Zealand the phosphate will have to compete against the imported guanos, and against agricultural limes produced by the burning of limestone; against these it is capable of holding its own, and of making progress.

Other districts in New Zealand may also contain unrecognised deposits of phosphate of lime. The discovery at Clarendon was followed by the discovery of a smaller deposit at Enfield, near Oamaru, but we have not heard much about this. Lime-stone occurs in quantity throughout New Zealand, chiefly in the South Island; and its surface, especially where eroded to any great extent, should be carefully examined for rock-phosphate. The phosphate had been overlooked at Clarendon for many years, and it may still be overlooked in some other locality. It is easily mistaken for limestone, flint, &c., according to the variety met with. It would be well worth while to submit to qualitative chemical analysis any peculiar-looking rock-fragment