USGS Professional Paper 144 pp 146-156

In the preceding pages and on the accompanying maps are presented in considerable detail the facts so far as they have been ascertained regarding the rocks of the district and the occurrence of the ores. The different theories of the way in which the ores were deposited have also been discussed. It has been hoped that both facts and theories would be of assistance in the search for new ore deposits.

This geologic report might perhaps wisely stop here and leave the practical application of its contents to be made by the mine operators of the district. Certain features that bear on the search for new ore, however, will be pointed out, but readers may draw their own conclusions, which may differ more or less from those presented in the following pages.

The advisability of searching for new ore bodies in the lodes is to be measured by the economic record of the ore bodies already mined and by the probability that similar bodies remain which can be found at reasonable cost. To the end of 1925 the mines of the district had paid in dividends about $290,000,000. It is safe to say that before the present known ore bodies are mined out and the companies liquidated, the dividends will be at least $400,000,000. There is no close record of capital expenditures, but from available data they are estimated at $150,000,000. It is apparent, therefore, that the district has yielded a fair profit, probably better than the average for mining districts. A large part of the dividends have come from operations on a few ore shoots, and the capital expenditures in developing these shoots have usually been small, so that the profits on these particular operations have been large. Most of the capital expenditures, however, have been made in developments that have paid nothing. In this respect the Michigan copper district does not differ from most other districts.

Granted that several ore shoots have yielded large profits, how likely is it that there are similar shoots in the district as yet undiscovered? Most of the bedrock in the district is covered with glacial drift and therefore not open to inspection, yet although some deposits have been found by accident, most of the ore shoots have been discovered on the outcrops or by old Indian workings that doubtless started on outcrops. Once a shoot has been located, extensions have been traced. An inspection of the geologic map will give an idea of the amount of development work done outside the ore bodies and make it clear that the thoroughly prospected area is but a small part of the total. There seems no doubt that undiscovered ore shoots exist.

That these shoots are not easy to discover is indicated by past experience. At best a rather large expenditure must be made before the success of an enterprise can be established. This, however, should not discourage those in a position to undertake such operations, as it is a condition that now prevails in practically all mining districts. In other districts, as truly as in this, most deposits easy of discovery have already been found.

Much of the district is covered by glacial drift, which adds greatly to the expense of exploration everywhere and which over considerable areas, where it is several hundred feet thick, has to the present largely prohibited extensive exploration. Only the massive or otherwise resistant beds, such as the Greenstone flow, crop out prominently. The amygdaloids are relatively weak rocks and are ordinarily eroded somewhat below the traps and covered even where the traps are exposed. Though the lodes are rarely exposed, it has been mainly by exposures that the ore shoots have been found.

The copper deposits are distributed stratigraphically through several thousand feet of rock, and geographically the main productive portion occupies a belt 2 to 4 miles wide extending from Central to Victoria, a distance of about 75 miles, though the larger part of the production to date has come from the central portion of this belt, about 40 miles in length. This belt is among the largest mineralized areas in the world.

The production since 1845 has been comparable to that from the Butte district since 1850, which has come from a few square miles. Most of the copper produced in the Michigan district has come from a few large deposits, but these are widely scattered, and the ore forms but a very small part of the rock within the mineralized area. The location of the ore bodies, even though large, in so extensive an area is, of course, relatively difficult.

All the lode deposits are of relatively low grade. The average yield for all lode deposits from the Champion mine north has been about 26.9 pounds to the ton, and from 1906 to 1923 the average for all mines of the district was about 21.5 pounds to the ton. If the Calumet & Hecla conglomerate is excluded, the average yield from the northern amygdaloid lodes has been 20.8 pounds to the ton from the beginning and 19.6 pounds from 1906 to 1923. Although the ore shoots are unusually large and regular as compared with those in other districts, nevertheless it is evident that in some the grade is not far above the economic limit, and a slight change in grade or in operating conditions suffices to put it below that limit.

After an ore shoot has been found, it is necessary to open a large mass of the lode to determine the size of the shoot and grade of the ore before it can be known whether or not a mine can be developed. To determine these questions is more difficult and costly in this district than in most others. The lodes must be opened by underground workings extending for thousands of feet, and mill tests must be made on the rock. Owing to the irregular distribution of the copper, no method of sampling that has been developed gives even a safe approximation of the grade of the ore. When the lode has been opened, those familiar with the ore can make a rough estimate of the copper content. If it is rich, the operators are warranted in going forward, but if it approaches the lower limit, only extensive mill tests will determine whether it is commercial or not.

In the past large sums have been expended in the development of shoots that proved to be below commercial grade, and it seems inevitable that this experience will be repeated in the future. It is not necessary, however, to erect an expensive mill in advance of proving a deposit, as has been done in some places in the past, for it is usually possible to make tests in existing mills, and in recent years this has been the custom.

The belief is sometimes expressed that if ore of somewhat lower grade could be successfully mined, immense deposits would be available, but the increase would probably not be as great as is thought by some. Some ore shoots grade at the margins rather gradually into leaner and leaner ore, but it is far more common to go within a relatively short distance from profitable to hopelessly unprofitable ground. Within an ore shoot considerably more ground could be taken if the commercial grade were lowered.

It is perhaps common to think of high-grade deposits as necessarily small and of low-grade deposits as likely to be large. Whether or not there is any justification for such an idea as a general principle, it does not seem to hold good for the Michigan lode deposits. None of the great ore shoots have been mined out, nor are different shoots mined to the same extent, and therefore no final comparison is possible. Nevertheless there is a strong indication, as is suggested by the following table, that the larger shoots as measured by their content of copper, are consistently of the higher grade. The richer shoots have been most profitable and therefore most developed; if the leaner shoots had had the same amount of development they would probably show better than they do.

Shoot	Copper(pounds)		Dividends
	Total	Per ton	Total	Per pound of copper (cents)
Calumet & Hecla conglomerate	3,375,000,000	49.07	148,700,000	4.41
Kearsarge	1,177,000,000	19.27	50,880,000	4.32
Pewabic	873,700,000	22.11	29,242,500	3.35
Baltic	873,000,000	^a26.50	43,004,000	4.93
Osceola	416,400,000	16.75	14,700,000	3.55
Isle Royale	210,800,000	16.39	2,550,000	1.21
Atlantic	142,800,000	13.55	990,000	0.69
Mass	50,000,000	^b15	-------------
Superior	31,000,000	^b19	^c649,000	2.10
Victoria	20,000,000	^b12	-------------
White Pine	18,000,000	20.3	33,337.50	0.18
Winona	17,000,000	^b14	-------------

a The high grade of the Baltic lode is due in part to sorting. b Approximate. c Includes final liquidation dividend, 1925.

It is apparent that the relation between grade and size of deposit as indicated by mining to date is not entirely consistent, and of course the actual size is undetermined. Nevertheless there is an unmistakable indication that the larger deposits are the higher in grade. If, then, a low-grade ore is encountered in prospecting, it should be developed to make sure that the openings are not in a poor spot in or on the margin of a richer shoot, but when it is once established that the shoot is of low- grade, there is little reason to believe that it will make up in size what it lacks in richness.

The size and grade of a lode deposit necessary for success in operation will of course vary with conditions. An inspection of the preceding table will show that in the past the profits have been roughly in proportion to the amount of ore mined and the grade of the deposits. As a rule, the lode deposits that have produced more than 100,000,000 pounds have been consistently profitable and those that have produced less have been rather consistently unprofitable, though the Superior and White Pine deposits; because of rather rich ore and other favorable conditions, have made earnings from a much smaller production. The lowest grade of ore that has generally been profitable in the past seems to be about 15 pounds to the ton; though the Atlantic shoot, which yielded a profit, averaged somewhat below that. It hardly need be said that relatively small mines located in rich parts of the large shoots, such as the South Kearsarge and Wolverine mines in the Kearsarge shoot, have been very profitable.

The profits from operations depend on many factors that will not be discussed here. The most detailed published discussion of the subject is one by Denton,⁶³ who reaches the following conclusion:

It seems likely therefore that the minimum requirements for a profitable mine in one of our amygdaloids are, approximately, that 50 per cent of the lode must produce around 20 pounds per square foot of lode mined, concentrated into 1 ton of stamp rock. The mine one hopes for must show at least 60 per cent of the lode area producing 30 pounds of copper and 1 ton of stamp rock per square foot.

In other words, the minimum requirements where the rock is not sorted are that in one-half of the developed ground the lode must be 11 feet thick and yield 20 pounds of copper to the ton, and in the "mine one hopes for" 60 per cent of the developed ground must have a lode thickness of 11 feet and yield 30 pounds to to{sic} the ton. Where the rock is sorted the lode might be thicker and correspondingly leaner.

It is hardly necessary to say that an ore shoot must be considerably developed before any clear idea of the size and grade can be gathered. Where extensive development has been made on reasonably encouraging ground but failed to develop a mine, the resulting loss may be diminished by mining some of the best ground opened.

Fissures have yielded a comparatively small proportion of the copper and of the dividends for the district. In dividends per pound of copper, however, they compare favorably with lode deposits. The fissures that have proved profitable have been those that yielded mainly mass copper and, with the exception o£ the Mass fissure of the Ahmeek mine, were rich and profitable from the surface. In deposits of this class it was not necessary to open extensively in advance of profitable extraction, and the fissures that proved profitable were so almost from the start of operations. The fissures commonly contain a considerable amount of stamp rock, but those in which the copper was mainly in stamp rock have not been profitable. There are numerous fissures on the Keweenaw anticline that have been prospected but little or not at all, and the incentive for fissure exploration and is much the same as that for lode exploration.

Certain geologic conditions that have proved favorable to the formation of deposits may be briefly reviewed here in their bearing or, exploration.

Amygdaloid lodes have been separated into four classes -- cellular, cellular-coalescing, fragmental, and scoriaceous. The characteristics of these classes have been fully discussed in preceding sections and need not be restated here.

No mines or encouraging prospects have been developed in typical cellular amygdaloids, which appear unqualifiedly unfavorable. This applies equally to all cellular rock, whether the top of the flow is of that character throughout its extent or whether it is fragmental in places and cellular in places.

The fact that the lode is fragmental in one place increases the probability that it may be fragmental in other places. All but two of the largely productive amygdaloids of the district and most of those that have given some encouragement are of the fragmental type. There seems no doubt that this is the most favorable type. The fragmental amygdaloids form not more than 10 to 15 per cent of the total, and cellular amygdaloids make up most of the remainder. It is evident, therefore, that the physical character of the amygdaloids affords criteria by which a large proportion can be eliminated.

A considerable proportion of the copper from the Pewabic amygdaloid lodes of the Quincy mine has come from cellular-coalescing rock, though the lodes are in part fragmental. There is no doubt that valuable deposits can and do occur in the cellular-coalescing lodes.

The Ashbed is the only scoriaceous lode in which shoots have been opened. The Atlantic mine is the only deposit in the Ashbed that has been profitable, and that not largely so. It is evidently possible for the lodes of the scoriaceous type to be mineralized to the extent of containing profitable shoots, but they are pretty clearly on the border line between favorable and unfavorable.

The influence of hematite in causing the precipitation of metallic copper indicates that highly oxidized lodes are the most favorable for ore deposition. The same influences that tend to form fragment al lodes seem to cause high oxidation, so that in the main the lodes that are physically favorable are also chemically favorable. It may be pointed out, therefore, that cellular amygdaloid that is highly oxidized may be more likely to pass into fragmental amygdaloid than that which is poorly oxidized.

Only two of the several conglomerates of the district have been shown to be extensively mineralized, namely, the Calumet & Hecla, and the Allouez. Both, where mineralized, are moderately coarse, and the Calumet & Hecla conglomerate, where it thins and changes to sandstone, quickly decreases in copper content.. It appears that a moderately coarse conglomerate has the permeability requisite for forming an ore deposit. It seems probable that a moderately thin conglomerate that pinches out in places along the strike but is continuous down the dip is more favorable than a thick conglomerate continuous for long stretches that gives no opportunity for convergence of rising solutions. All the conglomerates have relatively abundant ferric oxide and are in this respect apparently chemically favorable to copper deposition.

The sandstone of the Nonesuch formation at the White Pine mine and neighboring prospects is the only sandstone known to be strongly mineralized and that is mineralized only relatively near to fissures. Sandstone therefore does not seem to be favorable to mineralization, though it is of course more favorable than shale.

A detailed study of the mineralogy and paragenesis of minerals has been a feature of all the more comprehensive geologic studies of the district, starting with Pumpelly's work and continuing through the present investigation. One of the objects of these mineralogic studies has been to find, if possible, minerals that would serve as indications of either the presence or the absence of copper in their neighborhood. It may be said at the start that these studies have not developed any very positive aids in either the general exploration stages or the later development and mining stages.

Mineralogic features, like textural features, are largely of local occurrence, and the best of ore shoots are known to contain patches of minerals that are regarded as unfavorable indications of copper; likewise many amygdaloids which so far as known contain no ore shoots may for long distances carry mineral combinations that are characteristic of some of the big ore shoots in the district. In. the later stages of development observation of the copper itself is a better indication of the grade of the ground than a study of the gangue minerals.

The principal minerals that are regarded favorably as indications of copper in amygdaloids are the quartz-pumpellyite-epidote combination which is associated with copper, especially in the Isle Royale, Baltic, Pewabic, and Evergreen lodes. However, many impermeable cellular amygdaloids that show alteration of this type are encountered in diamond drilling and crosscutting. The presence of these minerals should not lead to more extensive examination by underground openings unless other features, such as favorable physical character of rock or presence of copper, give additional cause for encouragement.

The rock bleached through the removal of iron, on the other hand, is much more rare and, so far as known, is invariably accompanied by copper. Furthermore, there is a distinct feeling that this bleached rock is indicative of a rather intense degree of mineralization, so that if a drill hole or a crosscut encounters a little copper surrounded by bleached rock, it is regarded more favorably than a similar quantity of copper without the bleached rock.

Red feldspar, prehnite, and datolite are to be regarded as more favorable than otherwise. The prehnite usually has some fine copper associated with it though prehnitized areas in the Osceola lode are rather poor. In general a fragmental lode showing these minerals, with perhaps some epidote and pumpellyite, is to be regarded more favorably than a lode of the same type with nothing but calcite in the interfragmental spaces. What probably amounts to the same thing is that the greater the variety of minerals the more favorable the appearance of the lode. Most of the commercial lodes have a greater variety of minerals than the average amygdaloid:

Calcite and chlorite are rather indecisive indicators. The intense chloritization in some places, such as adjacent to fissures in the Kearsarge lode and in the shattered areas near the Keweenaw fault, would seem to be an unfavorable sign.

Laumontite is generally regarded unfavorably. However, there are local patches of lean laumontitized rock in the best ore shoots.

The value of copper itself as an indication of an ore shoot needs some discussion. If copper is present in a drill core from an amygdaloid, it is of course always encouraging, but it is more encouraging under certain conditions than under others. The most favorable mode of occurrence is in association with rock bleached by the removal of iron in a fragmental amygdaloid. Copper associated with the pumpellyite-quartz-epidote combination is also favorable, especially if in a fragmental lode.

Copper in a fragmental lode leads to interest in the lode at that point, whereas copper in some quantity in a cellular lode encourages examination of that lode elsewhere to see if it changes into one of more favorable character, such as a coalescing cellular lode or a fragmental lode. Very fine copper in prehnite or calcite is pretty common and not very encouraging. Copper in small seams in amygdaloids of any type is in general of little interest.

The variety of introduced minerals in conglomerates is much less than in amygdaloids, but in general the same principles apply to the use of mineralogy in exploring and developing rock of both types. Many of the conglomerates seen in drill cores have practically none of these minerals, with the exception of a little calcite. As in the amygdaloids, the occurrence of rock bleached by iron removal as a rule probably indicates copper mineralization. Likewise, it is probable that a greater variety of secondary minerals in a conglomerate is a better indication than calcite alone.

Little if any choice between two fissures can be made on the basis of mineralogy. The chief gangue mineral in the Mass fissure at the Ahmeek mine is calcite. The Owl Creek fissure at the Copper Falls mine in places had an abundance of datolite. Prehnite is common in many of the fissures of Keweenaw County. Some of the barren "crossings" in the lodes of the district are said to be mainly calcite. In many places the Mass fissure at the Ahmeek mine is merely a chlorite seam. In general, other things being equal, a fissure with a variety of gangue minerals is more encouraging than one with only a few.

In summary, although mineralogy is of value in exploration, its use must be accompanied by an understanding of its limitations. Some of the limitations are indicated. The best sign of copper is copper itself and the occurrence of copper in favorable lode rock is a much better sign than all the gangue minerals without copper in unfavorable rock. In development and mining mineralogy is of less use. In these operations the copper itself as seen in the openings is the best guide.

In several of the lodes it is pretty clear that the ore-forming solutions have traveled upward for long distances in the lodes themselves. It may thus be assumed that a long downward extension of permeable lode is essential to the formation of an ore shoot. It might be supposed that long downward extension would accompany long lateral extension of flows and that therefore the relatively thick flows extending for long distances along the strike would be most favorable. There is some support for this idea in the fact that the Kearsarge, Osceola, and Baltic are all relatively thick flows and that the first two are known for long distances along the strike. The individual flows of the Pewabic amygdaloid are relatively thin and apparently not continuous for long distances. However, if the flows came from the central part of the basin, any that reach the present outcrop may be assumed to extend a long distance down the dip. It is questionable, therefore, whether thickness of flow should be given any weight in the matter of favorability.

Barriers in the lode itself. - There are two general conditions in the lodes themselves favorable to concentration of solutions - (a) a lode that is prevailingly impermeable but contains permeable areas having a long downward extension; (b) a lode that is prevailingly permeable but contains bars of impermeable rock so placed as to cause a concentration of rising solutions. The Calumet & Hecla conglomerate and the Kearsarge aniygdaloid are examples of the first type, and the Osceola lode of the second. Recognition of the first type should be of help in prospecting. In the second type, where the barrier in the lode is relatively small, it is probably quite as easy to find the ore shoot itself as it is to find the barrier. In the development of an ore shoot once found, however, the recognition of the barrier should be of decided help.

Folds. - It would be expected that solutions rising along a lode would tend to concentrate near the crests of the anticlinal folds that extend down the dip transverse to the general strike of the rocks. The presence of ore shoots on the Allouez, Baltic, Winona, Mass, and Michigan anticlines lends support to this idea. On the other hand, there are several deposits, such as the Calumet & Hecla conglomerate, Osceola, and Quincy, that are not on folds, and the Isle Royale and Forest ("Victoria") are on synclines.

In the shoots studied in detail that are on anticlines the distribution of copper is more closely related to the character of the lode rock than to the crest of the anticline. The positions of the anticlines are well known, and if the idea proves to have any merit it can be easily applied.

Faults. - A fault offsetting a lode and making an angle with the dip of the lode would be a barrier that would tend to concentrate rising solutions the same as impermeable rock. Many faults are known that have sufficient throw to produce this effect, and there are doubtless many small unknown faults that might have this influence.

The ore shoots in several fissures are at or near the intersection with thick, well-oxidized amygdaloids or conglomerates, and. around the Keweenaw anticline they occur under the Allouez "slide," which probably acted as a barrier. These relations should be kept in mind in the prospecting of fissures.

The ore deposits in general are distributed geologically from a horizon near the lowest exposed part of the series to one well toward the top of the portion where flows predominate. The deposits in the Nonesuch formation are much higher but so far as known are present only near the Porcupine Mountain dome. There seems to be no very systematic arrangement in the geologic horizons at which the deposits occur. Near the central portion of the district, in a relatively short stretch from Quincy to Baltic, lodes occur from the highest to the lowest and at intermediate horizons. To the north the largely productive lodes are at intermediate horizons, though the Ashbed high in the series has been productive well to the north, at Copper Falls. In the south end of the district the productive lodes have been mainly near No. 8 conglomerate, at an intermediate horizon. From the distribution of the known deposits there seems little reason for favoring one horizon in the series over another.

Geographically the most productive part of the district, from Champion to Mohawk, is centrally located. Lode production greatly predominates in this central portion. North of the more highly productive area there has been considerable output from fissures, and south of it from both lodes and fissures. There is some reason to think that mineralization decreases in amount north and south of the central area, but it is not to be supposed that the commercial limits are those indicated by the present profitable lode mines.

The assertion frequently made that only one deposit occurs in any section across the mineralized belt is not wholly supported by the facts. The Calumet & Hecla conglomerate and Osceola shoots in part overlap. The Allouez conglomerate is mineralized and was mined above the Kearsarge shoot. The Atlantic in part overlies the Isle Royale and Superior. The Quincy has mined several lodes in the same section, and in the south end of the district the same is true of operations on the Evergreen and succeeding lodes. There seems no good reason for considering that an area is distinctly less promising because an ore shoot has already been developed in the cross section of the belt in which the area lies or that a section in which no ore shoot has been developed is particularly promising for that reason alone.

In no lode, up to the present time, have two widely separated profitable ore shoots been developed. Widely separated shoots have been developed in the Ashbed, in the Allouez conglomerate, and in the Evergreen and succeeding lodes of that series, but not more than one of those in a given lode has been profitable. There seems, then, to be no basis for supposing that a horizon at which productive deposits have been found in one part of the district is a particularly favorable horizon for prospecting in another part far distant, nor for prospecting a well-known and productive lode at a place far from an area where it is known to be productive, unless at this place it shows favorable character and signs of mineralization. Prospecting of well-known lodes at a distance from areas where they are known to be productive has been pretty thoroughly tried, and to the present time it has consistently resulted in failure. There seems no reason, however, why a lode should not contain more than one ore shoot, and if it is favorable in character at a distance from the known shoot, it should be considered as attentively as other equally favorable lodes. If it is physically unfavorable, on the other hand, it deserves no more attention than other unfavorable lodes.

In short, a given lode at a given place should be treated according to the local indications, and no money and effort should be expended on the basis of what the same lode may contain 20 miles away.

All known important deposits appear to have reached the surface and to have been as rich at some places along the outcrop as at any greater depth. (See p. 112.) It would seem, therefore, that in prospecting the chance of striking a shoot at the richest point is probably as good at one depth as at another within practical limits. The depth at which general exploration should be carried on should be governed by the cost at the particular place considered. In new territory relatively shallow depth will usually be cheapest. Near old mines deeper exploration from existing openings may be just as cheap, and the deep openings may develop much less water and thus cause less future expense in pumping in case the prospect is abandoned. It is of course to be recognized that exploration for known or supposed shoots entering a property at depth constitutes a special case.

Exploration in the Copper Range has now been in progress for more than 75 years and in different places has reached very different stages in its progress toward what may be regarded as complete exploration. For the portion of the range in Wisconsin the Wisconsin Geological Survey has now in progress the outlining of the broader geologic relations. For small areas in Michigan very detailed information is available. All stages between those extremes are to be found.

The problem to be solved in any given area and the method of attacking it depend on the information already in hand. In an area of which little is known the effective method is to proceed from the more general type of information to the more detailed. Lode exploration may be efficiently conducted in the following general order: (1) Geologic reconnaissance; (2) location of favorable lodes; (3) search for ore shoots; (4) development of prospects; (5) development of known ore bodies.

Geologic reconnaissance. - For most of the copper belt of Michigan the early stage has been passed, and the general distribution and relations of the rocks are known. Near the end of Keweenaw Point and toward the Wisconsin boundary there is still something to be gained from work of this type.

Location of favorable lodes. --- With the general location and geologic relations determined, the next step is to find what are regarded as favorable lodes and conversely to determine unfavorable lodes.

For the area between Victoria and Breakfast Lake the accompanying geologic maps and sections give the character of the lodes and their position so far as known. For some areas this information is rather detailed, but for others it is very slight. It is evident that much remains to be done in this stage of exploration.

Search for ore shoots. - As is well known, no lode is mineralized to a commercial degree over more than a small part of its extent. Moreover, numerous beds that have all the properties that are regarded as favorable to mineralization are not known to contain minable ore at any place, though no lode has been thoroughly tested throughout. It is clear that the copper occurs in distinct shoots, only the larger and richer of which are of commercial importance.

It is always possible that a shoot may be discovered at any stage of exploration, and as a matter of fact most of those now known have been located in the early stages or entirely by chance. The probability of locating additional shoots by finding outcrops containing copper or by following up old Indian diggings is not great. Ore shoots are likely to be found in the future only by systematic search, implying an exploration program and the use of methods materially different from those that were cheapest in earlier stages.

Development of prospects. - Once a stretch of lode that contains an encouraging amount of copper is located, it must be developed to determine whether or not it is of a size and grade to make a mine. At this stage in particular an understanding of the behavior of the shoots in the developed mines is likely to be of assistance.

Exploration of ore shoots. -- The mining of the great ore bodies, if it is to be carried out efficiently, must be planned far ahead of actual operation. The determination of the probable position and extent of a known ore shoot in undeveloped ground is an important function of exploration.

The several methods of exploration that have been commonly applied in the region, usually in the order named, are (1) examination and mapping of surface exposures; (2) trenching or digging test pits where the overburden is shallow; (3) diamond drilling; (4) underground openings. Other methods have been tried on a far less extensive scale.

Examination of surface. - To get all the information that is possible from an examination of the surface is of course the first step in exploration work at any stage. It was early applied and resulted in outlining the broader features of the geology which have long been known. This method is particularly useful in the reconnaissance stages of exploration but may also be useful in later stages. Considerable detailed information may thus be obtained for small areas, especially along stretches of lake shore and along river channels. Elsewhere only the more resistant beds are commonly exposed, and the amygdaloids, which are weak and easily eroded, usually occupy relatively depressed areas and are therefore commonly covered.

Trenches and pits. - The results obtained by digging trenches and pits in exploration are of course dependent on depth and character of overburden and on water conditions. In general this method is practicable only in areas covered by a few feet of overburden and is not applicable to swampy areas even where the overburden is thin. Wherever it can be effectively applied, this method is useful in all the earlier stages of exploration - namely, geologic reconnaissance, location of favorable lodes, and search for ore shoots.

Diamond drilling. - Diamond drilling has been employed in exploration in the copper country since 1882 and has probably furnished more detailed geologic information than all other methods combined. It is highly effective in determining the character of flows for general correlation, and where skillfully conducted it has proved very useful in determining the character of lodes. The medium-sized 1 1/8-inch core from the "A" bit is much more informing than the 7/8-inch core from the "E" bit and is well worth the difference in cost. Although diamond drilling is the most efficient method of getting information as to the kind of rocks and the character and position of lodes where outcrops or surface trenching will not furnish this information, the diamond drill has proved of rather doubtful service in the search for ore shoots. Owing to the very irregular distribution of copper in the average amygdaloid lode, a drill-core sample is likely to be equally misleading whether it contains copper or not, though the presence of copper should, of course, always be regarded as encouraging. Cores sufficiently encouraging to cause the sinking of exploratory shafts have been obtained at several places, as Mandan, Ojibway, Mayflower-Old Colony, and St. Louis, where disappointing results were encountered when the lodes were opened. The Lake lode, on the other hand, was located by diamond drilling and proved sufficiently mineralized to lead to rather extensive development. At the New Arcadian mine the lode that showed best in the drilling proved disappointing, but a near-by lode that was opened in the underground exploration has been extensively developed. Diamond drilling is thus a far less reliable guide in the search for ore shoots than trenching or underground openings, methods by which it is advisable, where indications warrant the expense, to expose considerable portions of the potential lode. Obviously, however, the information as to copper content obtained by diamond drilling is not to be neglected. Drill-core samples are more nearly representative of conglomerates than of amygdaloids, because the copper is more evenly distributed in the former rock than in the latter. The uniformity of distribution and the consequent reliability of drill-core samples is naturally even greater for finer sediments; the cores have been shown to give the copper content of the Nonesuch sandstone and shale, for example, with a fair degree of accuracy.

The diamond drill has been used rather extensively in some mines to locate copper ground in known lodes. It has been used most in the Quincy mine, where the lode is of the coalescing type, and for that type it is said to be effective. It was also used in the Osceola mine to locate "foot lode" copper, but here it produced rather indifferent results and was given up. In the average amygdaloid lode the advisability of its use for this purpose seems open to question.

Foot for foot, diamond drilling is, of course, much cheaper than underground opening. Probably at least 4 feet of drilling can be done at the cost of 1 foot of underground opening. In cross-sectioning inclined beds the ratio is even more favorable to the drill, which can cut the beds at right angles and traverse a given stratigraphic thickness with a minimum footage, as a crosscut can rarely do. It may thus be possible to section a lode or series of lodes several times with the diamond drill at the same cost as a single section with an underground opening.

Churn drilling. - Churn drilling, so far as known has been used only at the Laurium property. The records of this work do not indicate that it is effective in giving any type of information in this district, and drill cuttings must obviously be far less informing than a core. The churn drill might be effective in drilling through overburden to bedrock.

Underground opening. - Underground opening, of course, gives, foot for foot, the most satisfactory information of all kinds. For general geologic information or the location of favorable lodes it is ordinarily too expensive. In the search for ore shoots where trenching is not practicable it is probably the most effective method. In the development of prospects it is the only effective means of determining the position of the ore shoot and the copper content of the ground.

Other methods. - The dip needle has been employed by the Wisconsin Geological Survey in working out the general distribution and structure of the copper-bearing rocks. This method has been but relatively little employed in the Michigan copper district, and for much of the area more detailed data are already available than are likely to result from dip-needle work alone.

For certain problems, even in the more intensively developed areas, the dip needle will furnish geologic data at a lower cost than most other methods. In the less developed areas it has a larger field of usefulness. The possibilities of the use of the dip needle are discussed in the section on geophysical methods (pp. 156 168). Other geophysical methods have been but little used in the Copper Range. Electrical methods have been tried as outlined on pages 158-160 but have given as yet no very encouraging results.

The results attained with a given amount of money vary considerably with the skill used in choosing and applying the different means of exploration, even when the simplest methods are employed.

A careful determination of the type and grain of the traps may give considerable information on the thickness and correlation of beds that are only slightly exposed and thus save more expensive work by restricting trenching and drilling to the vicinity of flow tops.

In trenching across the strike of flows, if the grain of the rock is determined, it is possible to judge the approximate thickness of many flows and to avoid continuous trenching over the central portion of such flows.

Diamond drilling has been pretty thoroughly developed and systematized in the district, but there are. certain features that may well be emphasized.

In making cross sections it is the general practice to drill the section across the supposed strike and, unless there are special reasons for doing otherwise, at nearly right angles to the supposed dip of the beds. This of course gives the maximum section for a. given footage. It is worth going; to considerable trouble, if necessary, to determine the strike of the beds before starting an extensive cross section. The purpose of a cross section is to determine not only the character of the beds but their attitude, and for this purpose a correlation between two holes is necessary. It is desirable, therefore, to find a characteristic bed that will permit correlation near the bottom of the hole farthest down the dip and to cut this same bed near the top of the next hole up the dip. It is well worth while to carry a hole some distance beyond what would otherwise be the most economical limit in length or to stop it short of that limit if such a correlation can be effected. A sedimentary bed is usually the best for correlation, but certain characteristic flows are satisfactory.

If beds of favorable character are encountered near the bottom of a hole, it is worth while to cut these in the next hole to determine their character at another point. To accomplish this, a hole may be stopped short of the most economic length after passing through favorable amygdaloids in order to start the next hole near the outcrop of those beds and cut them again.

Both the "A" bit, with 1 1/8-inch core, and the "E" bit, with 7/8-inch core, have been used in the district. The "A" bit has given a far better core recovery, especially of the amygdaloids, and therefore a better idea of the beds, and this advantage is well worth the difference in cost of operation. Larger bits have been used, but except under special conditions they are probably not justified by the additional information obtained.

A careful record should be made of every hole at the time of drilling, but in addition the core should be preserved, so that it can be reexamined when desired. A core may be stored at a cost of but a few cents a foot, and there is no justification for not preserving it. A few suggestions resulting from the examination of many thousand feet of core may not be out of place.

Cores are usually stored in wooden boxes ranging from 5 to 9 feet in length and containing from five to eight or nine rows of core each. The rows of core are commonly separated by wooden dividers. The size of box is not material, but pine lumber, even of poor quality, is preferable to even high-grade hemlock. Sheet zinc may be used for dividers in place of wood strips. This saves some space, and the cost per foot of core is about the same.

Marking of the depths within the box and marking of the box are highly important. For marking depths, 1-inch wooden blocks the width of the core should be used. The depth should be clearly marked in pencil on the top and one side of the block. A block should be placed at the end of each pull. It is convenient if the beginning of the box is marked with an arrow, and it is advisable to follow the general custom of placing the cores so that they are to be read from left to right as a book is read. Red marks on the dividing strips opposite places where copper is found in the core are helpful. The box, not the cover, should be marked both inside and outside with the name of the property, the number of the hole, and the depth. Metal tags with the numbers and letters stamped on are durable, but the box itself should be marked in addition.

Cores should be stored in a dry place and piled well above the ground. If left out of doors or in a leaky building, they might as well be thrown out at once, as the boxes will soon be decayed. (See pl. 54, B.) Some years ago the Michigan Geological Survey, recognizing that the matter was one of public interest, undertook to preserve such cores as were not being cared for by the companies. Unfortunately this effort was not continued, and those already collected were not given proper care.

Depth. - The first question to determine in exploration is the most favorable depth. As an ore shoot is as likely to be found at one depth as another (see p. 112), the depth of exploration depends on the cost, and in most places the most economical depth would probably be shallow.

Horizontal versus vertical exploration. - In many of the explorations in the district, shafts have been sunk to a depth of 1,000 to 2,000 feet, and the vertical element of exploration has been emphasized, as contrasted with exploration along the lodes. Whether the chance of encountering an ore shoot will be greatest by opening down the dip of a lode or along the strike will depend on the attitude of the ore shoot. If the shoot extends directly down the dip, exploration in that direction will be least effective and drifting along the lode most effective, but if the shoot is horizontal the reverse will be true. Most of the known shoots trend more nearly with the dip of the lodes than with the strike, and therefore, in general, drifting should be preferable to sinking. Drifting, moreover, has the advantage of being cheaper; it also affords the easier method of following an irregular lode and of testing any lode by crosscuts.

If encouraging results are not encountered at one level, it is very questionable whether drifting at another level is justified in general prospecting, but if the extension of a known shoot is being sought, drifting at another level may be desirable.

Examples are cited in the discussion of ore shoots where a lode is poor in stretches near the surface and richer at depth, but as a rule drifting along the lode would have encountered the rich shoots with less effort than opening down the lode.

If an encouraging amount of copper is encountered, it is of course worth while to open the shoot down the dip as well as along the strike, but it is desirable to follow the shoot along the strike to its limits on each level opened in order to get the horizontal extent and probable attitude as early as possible. If a lode is rich, it is common practice to run levels in it during development, at the regular operating intervals of 125 to 150 feet. If the deposit is of doubtful grade or size perhaps the more common and better practice is to open it only at considerably greater intervals; the probable extent and grade are thus determined at a lower cost.

Transverse versus longitudinal exploration. - In past exploration two practices have been more or less consistently followed-the examination of a lode or lodes known to be mineralized elsewhere in the district, to the exclusion of the other lodes; and a preliminary cross sectioning of all the lodes in an area by diamond drill, trench, or crosscut and the further examination of any encouraging lodes that may be found. The information already available for much of the district gives a general idea of the character of the beds, from which it can be judged what method is likely to give the best results. Crosscutting would usually be employed where it is desired to investigate several adjacent lodes that are known to be of favorable character. If, on the other hand, diamond drilling has shown a promising lode in a series of unfavorable lodes the promising lode would of course be best examined by itself.

Prospecting fissures. -- It has been pointed out that many of the known fissure deposits occur at the intersection of fissures with strong lodes and also that they may be under barriers like the Allouez "slide." These, then, are the places to be specially examined by the method that local conditions render easiest.

Where fissures are closely spaced it may be advantageous to drift on a strong amygdaloid across a number of such fissures, further developing any that show promise. Where they are widely spaced it will probably be advantageous to examine each fissure separately.

Equipment for prospecting. - It is obvious that no prospect in its earliest stages of development justifies the assumption that it will be amine. Therefore, until a prospect is pretty well proved the equipment and the openings should be the cheapest that can be effectively employed. Certainly the building of a mill should wait till the ore for it is fully assured. Such a policy may lead to slight losses in the relatively few prospects that develop into mines, but the owners of such properties can stand the loss.

The present report is but one of a series that have been made during the last 75 years. Each of the reports recorded the available facts concerning the geology of the district and expressed the opinions of the authors regarding the occurrence of the ores. With each succeeding report there has been a growing body of facts and a changing view regarding the occurrence of the ores. The change in view has been influenced by the accumulating of facts, by the general increase in knowledge of ore deposits, and of course by the views of the different individuals who have been engaged in the several investigations.

Each of the previous reports has been helpful in the development of the district, and it is hoped that this one will also be helpful. The authors of the report probably realize more keenly than others its shortcomings, and they are well aware how much remains to be done and how important the continuance of geologic work in the district is likely to be. Many of the problems can be solved only by a steady, persistent collection and correlation of data from year to year, and the necessity for this work will not cease till mining in the district is definitely abandoned.

One of the first necessities for a geologic study is an accurate base map. A modern topographic map has been prepared for the central part of the copper district in Michigan, but this should be extended over the district as a base for accurately recording geologic data.

The district presents numerous geologic problems that are still unsolved or only partly solved. Among these are the problem of the Keweenaw fault and the associated problems, such as the depth and attitude of the Keweenawan lavas beneath the "Eastern" sandstone, and whether or not they contain valuable copper deposits east of the Keweenaw fault. Solution of these problems will involve the study of the South Trap Range and its relation to the Copper Range.

The Porcupine Mountain area, the portion of the Copper Range westward from the Victoria mine to the Wisconsin boundary, and the area near the end of Keweenaw Point, in Keweenaw County, are but little known. Throughout the district, in fact, there are rather large areas of which little is known, as can be seen by inspection of the maps. Much can doubtless be done toward filling these gaps by magnetic surveys and other relatively cheap methods. This work should be done on a scale and with an accuracy suitable to use in mining development. Much trenching and other exploration was carried on in the early days, especially in Keweenaw County, of which there is no record. The location of these openings on an accurate base map would give considerable information as to what has already been done and as to localities where additional exploration is most promising. In this same area there is considerable to be done in mapping outcrops when a suitable base map is available.

It is already demonstrated that dip-needle mapping of the formations can be carried on to advantage, and there is a considerable field for this work in further mapping of the general geology.

The possibility of the development and application of geophysical methods to the search for ore in the district should be kept constantly in mind. The fact that they have not been notably helpful to the present time should not discourage the study of such methods.

Likewise there should be a continuing study of the ore occurrences with the purpose of discovering additional guides to the search for ore. Progress has been made in this direction over a period of 75 years, and it is clear that the end has not yet been reached.

To summarize, it may be stated that what is needed in this as in every other mining district is continuous geologic work with an accurate record and correlation of all data relating to the occurrence of ores and the steady and persistent attempt to apply new knowledge and new methods to the finding of new ore bodies and the exploration of those already known..

During the last few years considerable interest has been aroused in the application of geophysical methods to the search for economic minerals. Well-authenticated cases of the discovery of deposits so diverse as iron ore, copper sulphide deposits, gold veins, and oil pools, either directly or indirectly as a result of the use of these methods, are on record. The recent widespread interest in this subject is the outcome of the rapid development of the methods in the last decade, particularly in Germany, Sweden, and France.

One type of geophysical observation has long been known and used in the Lake Superior region. Every geologist in the iron districts regards the magnetic dip needle as an essential part of his equipment, and a "magnetic survey" is in by far the greater number of explorations one of the first steps. Curiously enough, however, the possibilities of magnetic observations, regarded as essential in geologic mapping in the iron districts of the adjacent counties, were practically ignored in the copper country. It is safe to say that the present high degree of completeness of knowledge of the stratigraphy in this district could have been obtained at an enormous saving had magnetic surveying been used in the explorations of the old type. Thousands of feet of diamond drilling and trenching could have been dispensed with, preliminary shafts and crosscuts, poking around to "find the lode," would have been unnecessary, many miles of expensive transit work on the surface could have been saved, and mines such as the Ahmeek and Mohawk, long undiscovered because of a simple curvature in the strike of the lode, would have been found as soon as anyone with a dip needle had taken the trouble to work along the strike from the original discovery. However, the results were obtained, even though at unnecessary expense; but it is to be hoped that the futile "cut and try" exploration will be indulged in less and less as time goes on and scientific principles of exploration are followed. The present problem is to decide what use to make of geophysical methods, in the light of present knowledge of the geology of the district. No very positive statements can be made in the following discussion of this subject. However, some experimental work has been done and consider able study has been given to the use of these methods by the geological department of the Calumet & Hecla Consolidated Copper Co., and we feel that we are in a position to say what there is to be said on the present status of geophysical methods as aids to geologic work in this district.

All geophysical methods depend upon a contrast in the physical properties of the mineral deposit itself with those of the surrounding rock, or upon the discovery or delineation by means of such contrasts of some geologic feature to which mineral deposits are related. Thus, magnetic deposits are more magnetic, massive chalcopyrite deposits are better conductors, and salt domes have a smaller gravitative attraction than the surrounding rock. Or a mineral body which of itself may have no outstanding physical property amenable to investigation by geophysical methods may be related to certain geologic features which could be outlined by such methods. Thus, oil pools are discovered by outlining salt domes, hematite bodies by their relations to magnetic dikes, or chalcopyrite deposits by their relations to a certain contact. The essential condition in all geophysical work is that there be a detectable contrast in certain physical properties of the feature that is being sought, be it the mineral deposit itself or some geologic structure or condition to which the deposit is related. The physical property upon which the method depends may be magnetic polarity, magnetic permeability, electrical polarity or conductivity, density, elasticity, or one of certain other properties that are used more rarely. The manner and degree in which this essential condition of contrast in physical properties is met by the rocks and ore bodies of this district will be brought out in the following paragraphs.

There are three radically different methods of conducting geophysical investigations. (1) If the physical properties of certain ore bodies or mineral deposits themselves have the necessary contrast with those of the country rock, that method can be selected which takes advantage of this contrast, and the work can be carried on with the direct object of locating the mineral deposits themselves. (2) It may be possible to take advantage of the physical properties of rocks associated with the mineral deposits, enabling structural features known to be related to the mineral deposits to be determined. (3) Geophysical methods may be used in general geologic mapping where the immediate objective is not so directly the location of mineral deposits as it is to build up a general geologic map which it is hoped will ultimately be of use in exploration. The following discussion will take up in order these three methods.

63 Denton, F. W., Development and extraction methods for Lake Superior copper deposits: Lake Superior Min. Inst. Bull., August, 1922.