US2200918A - Apparatus for sound recordation - Google Patents

Description

Patented May 14, 1940 UNITED, STATES PATENT OFFICE U 2,200,918 APPARATUS FOR- SOUNDYRECORDATIONV Orville M. Dunning, Glen Ridge, andCharles '1.

Jacobs. New Providence- Township,

Union County, N. J., assignors to Thomas A. Edison, Incorporated, West Orange, N. J., a corporation of New Jersey Application January 27, 1937, Serial No. 122,536

' 2 Claims.

. each time the sound is to be reproduced. Further, the invention especially contemplates recordation by impression upon a compound record-e. g., by compression and sideward displacement of a record body material by the action of a tool through an extensible record skin, without cutting or puncturing of the latter.

We are aware that direct impression recordation has been carried out on records which were temporarily softened (as by heat) during the recordation; but the necessity for any such temporary treatment is an intolerable. objection to a direct recordation process for a large number of purposes. We are also aware that direct impression recordation on a compound record in normal condition has been heretofore proposedbut in a system involving lateral, or side-to-side, soundrepresenting undulations in the grooves.

We have found, however, that excellent results and many special advantages accrue from a system, of direct impression recordation on a normal compound record, wherein the soundrepresenting groove undulations are verticali. e., are formed in the bottom of the groove. We have found, for example, that highly satisfactory recordings can be made with forces which are surprisingly small multiples of a force which may be exerted against the record. without detectable impairment of the recordings in repeated reproduction of excellent quality therefrom; this fact permits the production of a recording of great life (in terms of the important number of times reproduction) with moderate recording forces. Again, we have found both forces to be of a desirably lower order of magnitude than has proven practical in a lateral system otherwise of this general character.

We have found that the inherent slight elasticity of the compound recorda characteristic which tends sharply to militate against satisfactory recordation of lower amplitude (particularly high frequency) sounds in-a lateral system otherwise of this general character-produces far smaller such objectionable effects in our vertical recordation. Indeed, instead of finding it purely disadvantageous we believe it to be a material aid in the production of the desirably low force contrast abovementioned.

Another advantage in vertical impression on compound records is to be found in a reduced tendency of a given record skin to be disrupted and/or punctured during a vertical recording operation, which we believe "to be the result of the similarity of direction (i. e., the up and down. direction) of both the groove-forming and undulation-forming forces.

A satisfactorily low surface-noise-to-signal ratio in reproduction from the record seems easier I of obtainment with the vertical than with the lateral direct recording on normal compound records. We believe this results from a burnishing action of the recording tool on the relatively tough record skinan action peculiar to impression recordation on compound records and of course without analogy in cutting recordation-coupled with maintenance of this action at a high average level (as by the record-ward bias) on the reproducer-influencing bottom of the groove in vertical impression recordation, and with its maintenance only at a variable lower value (by the oscillatory side-to-side forces) on the reproducer-influencing sides of the groove in. lateral impression recordation. Indeed, in the latter case at low sound levels, when surface noise needs most to be reduced, the effective burnishing action is leastit being understood, of course, that although the bottom of the groove may be burnished in lateral impression recordation, the groove bottom does not in that case actively influence the reproducer.

It may be mentioned that in impression recordation generally an advantage of the use of vertical instead of lateral recordation concerns the production of echo or anticipation eflects in reproduction from the record; by these effects are meant the spurious addition at reduced volume, to the sounds recorded in the groove being traversed by the reproducing tool, of sounds recorded in the groove last traversed or next to be traversed by that tool. In reproduction from -any out recordings these effects are normally wholly absent unless the edge portions of the rooves actually overlap. With impression recordation, however, even in the absence of any groove overlapping, both effects (and in particular the "anticipation effect) may'occur to troublesome degrees; but with vertical impression recordation they are less likely to occur and easier to avoid. And of course there are certain alwaysrecognized advantages ofany vertical recordation over lateral, such as a greater possible numher of grooves per unit record dimension transverse to the grooves.

While we do not Wish to limit all aspects of our invention thereto, we have preferred to carry out our invention with a recording tool simply biased against the record, as distinguished from a tool whose mean position (relative to the rec- 0rd surface) is uniquely fixedin other words, in a system wherein depth of groove is proportional to record-ward biasing force. In vertical recordation by such a system which is arranged for cutting as distinguished from impression, there is usually encountered a serious difiiculty in the form of spurious oscillations of the recorder at a low frequency, accompanied by badly accentuated recordation at this and immediately neighboring frequencies; these oscillations are repeatedly occasioned by shocks and various other excitations, and their suppression frequently entails compromising, critical and otherwise troublesome practices. This difliculty we have found unexpectedly wholly absent from our vertical impression recordation on compound records.

In our preferred practice of simply biasing the tool against the record, we of course modulate the steady tool bias to effect the desired groove undulations. When a simple system of this character is employed in cutting, it is usually found that the change of groove depth per small unit change of force applied to the tool is reasonably contant at different depths, but that such deviation from constancy as does occur (except when quite extreme tool shapes are employed) usually causes an increase of this change with decreasing depths. The reverse of this effect-a sharp decrease of this change when the depth is decreased to minute values-is observed in our verticalimpression recordation on compound records, and is a corollary of our ability to reproduce from the record with a reasonable reproducer-biasing force without significant impairment of the record.

It is an object of our invention to provide methods and apparatus for direct impression recordation and reproduction, wherein the recordings will be characterized by excellent life.

It is an object to provide methods and apparatus for direct impression recordation with low applied forces.

It is an object to provide methods and apparatus for direct impression recordation with economical sound-representing power consumption.

It is another object to provide improved methods and apparatus for direct impression recordation on records in normal condition.

It is another object to provide methods and apparatus for direct impression recordation andreproduction with relatively low surface noise.

It is another object to provide methods and apparatus for direct impression recordation and reproduction with negligible echo and anticipa tion effects.

It is another object to provide methods and apparatus for bias-responsive vertical recording without spuriouslow frequency oscillations.

tain materials to peculiar advantage as such record skins.

It is another object to provide for an impression recording system an especially favorable record body material for employment with such record skins.

It is another object to provide for an impression recording system a generally improved compound record.

It is another object to provide generally improved methods and apparatus for impression recordation, especially for that which is direct. Other and allied objects will more fully appear from the following description and the appended claims. I

In the description reference is had to the accompanying drawing, of which Figure 1 is a partial cross-sectional view illustrating a typical record according to our invention and a typical cooperating stylus;

Figure 1a is a partial plan view of the same;

Figure 2 is a view, principally elevational but partly schematic, of such a record suitably supported and with a recorder for impression recordation thereon;

Figure 3, is a fractional cross-sectional view taken alongthe line 3-3 of Figure 1a;

Figure 4 is atypical curve of groove depth plotted against biasing force, for a typical embodiment of our invention; v

Figure 5 is a view, similar in nature to Figure 2 but illustrating reproduction from the record;

Figure 6 is a view similar to Figure 1 but illustrating a different typical stylus; and

Figure 7 is a partial cross-sectional view of an alternative form of record.

Reference may now be had to Figure 1, where- 1 in will be seen the impressible compound record ID. This may for example be of disc shape as indicated in Figure 1a; but of course there is intended no limitation of the broader aspects of or greater, caused to adhere, either by virtue of the characteristics of the layer l-2- or otherwise,

to the layer l2. Materials for these several elements of the record are hereinafter more specifically discussed. It may here be mentioned that the thicknesses of the elements have been exaggerated in' Figure 1 for the sake of clear illustration;

In Figure 1 there also appears the impressing tool or stylus 3 (which may be considered as forming a portion of the recorder I of Figure 2) arranged to engage the surface of the record If], the record of course being intended for continuous movement during recordation (for example in the direction indicated by the arrow).

One form of tool 3 which we have successfully' employed consists simply of a cylinder (say of .010" diameter) of any hard material (but desirably ofsapphire or the like), having its active end formed into a cone of 60 apex, and that apex rounded off in a spherical surface of small radius suohas .001; the axis of the cylinder and conical portion has been vertical (i. e., perpendicular to the record surface).

' Figure 2, on a smaller scale in which a showing of the laminations of the record has been omitted I cooperating with'the record. A convenient way in which to bias the tool against the record is to mount the entire. recorder for free movement toward the record, so that the biasing force becomes some function of the whole recorder weight-for example to pivot the main frame 2 of the recorder to a rigid member or bracket B as at the pivot P close to the record surface and,

typically, removed from the tool 3 in the direction from which the record moves. The oscillatory forces which modulate this record-ward bias of the tool may be imparted to the latter in any of a variety of manners; we have, however, illustrated an electro-mechanical force impartion through the medium of apiezo-electr'ic bending unit 4 (for example -of the form described in U. S. Patent NO.'1,802,782,'.iSSu6d April 28, 1931, to C. B. Sawyer) forming a portion of the recorder l and immediately carrying the tool 3.

This unit 4 maybe rigidly secured in and extend in cantilever fashion from a unit-base 5, and is arranged for'bending toward and away from the record. The unit base 5' may for example be. connected to an intermediate region of the recorder frame 2, and the unit 4 may extend therefrom in a direction away from the point P. The tool 3 may be secured underneath the unit at the free end of. the latter, as by means of a clamp 6 secured on the end of the unit.

Rigidity of the abovementioned connection of the unit base 5 to the frame 2 is one means of rendering the weight of the frame 2 effective for bias of. the tool 3; but we have preferred to make this connection a pivotal one (for unit movement toward and away from the record) as indicated by the pivot P, and to render the frame weight effective on the tool through resilient means rather than longitudinally through the unit 4. Thus the frame-2 may'extend above the unit 4 to beyond its free end; and to the clamp 6 at this end and to the frame thereabove may be secured the respective terminals of a spring 1,- through which the frame weight is made effective; .A finger 2b may extend downwardly at the end of the frame 2 and terminate below a lug 6b extending from the clamp 6, to limit downward pivoting of the unit 4 when the recorder is raised from the record.

It will be understood that upon the application of an alternating voltage-for example, soundrepresenting-to the leads 4a of the piezo-electric bending unit 4, the latter will bend oscillatorily and, in view of the inertia of the-unit base 5 and frame 2, will modulate the bias of the toolagainst the record. The record meanwhile being moved under the stylus-for example, rotated, with the entire recorder meanwhile being'slowly moved approximately radially of. the record in wellknown manner to render the path of tool over record a spiral one-the stylus will'impress a groove in the body material l2 through the skin l3; the groove depth will have a mean value regulable by the biasing, force 'on the tool, and

that .depth will be modulated in accordance with the alternating voltages applied to the unit 4.

A partial cross-sectional view of the record along the line '22 of Figure la appears as Figure 3, showing transversely a few grooves 8 such as made by the operation just described. By

way of example, the outer grooves may be considered as ofthe mean or unmodulated depth, and the grooves specially designated as 8a and 8b as of mutually converse peak-modulated depths.

3 More fully to understand the operation and power requirements of the system, we have found it helpful to accept the already known view of the record (in any vertical recording operation,

whether cutting or impressing) as an apparent spring acting on the tool in respect of up-anddown tool movement. This conception, while an' equivalent and not an actual one, is entirely warranted by the'substantial obedience of the tool during any such recordation to the law of a springi. e., a force is developed tending to rev store the tool to its mean vertical position, substantially in proportion to the displacement of the tool from that position (which obedience must obviously exist if the tool is at all faithpliance may betermed the dynamic recording compliance, since it is of course significant only while the system is in operation. In cutting recordation this dynamic compliance is usually and warrantably viewed as the sole essentialdynamic characteristic, and the recorder constants (for example in such a recorder as described, including the parameters of the piezoelectric unit 4) are accordingly determined simply for the development of. the maximum force on this compliance consistent with frequency characteristic and allied requirements. The actual power consumed by the recorder on such a case will be dissipated principally in various mechanical and electrical resistances which are incidentally present in the system but not immediately associated with the record. (It will be understood, of course, that the power under discussion is that consumedin modulating the groove; the record moving means is in any event the source of power for the simple formation of the groove).

In the impressing operation, however, definitely more modulating powerisrequired than in cutting, and this is evidence of an appreciable power-consuming component in the dynamic recording characteristics; this component may be termed the dynamic recording resistance.

. While it is negligible compared to the reactance in the cutting operation, we find it very appreciable relative to the compliance in our impressing operationit being typically the preponderate of the two characteristics over the frequency range from 500,to 1000 cycles upwardly. The

recorder constants are therefore desirably determined more nearly for delivery of maximum power to this dynamic resistance than merely for development of maximum force on the .dy-

Thus while this resistance has the effect of increasing the recordation power requirements, it has the desirable effect of opposing the mentioned oscillatory tendencies; and we accordingly prefer that it be, and find it in records such as hereinafter described inherently to be, present in a degree sufiicient to over-critically damp the objections in some cases.

oscillatory system whose reactive components are the dynamic recording compliance and the effective recorder mass.

Turning to a more detailed consideration of the record. characteristics, we may first point out the availability of a wide range of materials for the base II. Metal, such for example as steel, is eminently satisfactory, subject, however," to cost and weight objections in certain cases. Any of the normally hard plastic materials, such for example as phenol-resin compositions, are likewise satisfactory, subject, however, to cost Cardboard, preferably at least fairly stiff and hard or compact, has been employed satisfactorily and possibly best meets stringent cost and Weight requirements. In the selection'of a base material it is principally desirable that, when used in a simple embodiment as illustrated in Figure 1, it be sufficiently stiff to lie flat upon the turntable or other support and sufiiciently hard to insure that it will not, by itself yielding, absorb any appreciable part of the downward force of the tool 3.

While for thebody layer I2 any fairly inert (i. e.- non-elastic) yieldable material (preferably very approximately of'an order of hardness below mentioned) may be employed, it is of course con 'venient that it itself have a natural adhesion to the base I I and the skin l3. While certain waxes, compositions consisting largely of wax, or softened Celluloids or the like may be employed, we have found most desirable the use of a pitch, preferably of the asphalt variety, or a composition at, least principally of asph'alts. (In such an asphaltic composition it has been found advantageous but not indespensable to employ a small percentage for example of the order of 6%of a wax such as refined Montan wax, which serves as a blender for the several asphalts which may have been mixed, lowers the melting point of the composition, but tends to reduce change of the hardness over the range of temperatures normally encountered in the use of the records.

Particularly when a little wax has been so used, there may advantageously be added a few per cent.for example of the order of 3%of resin,

which tends slightly to harden the composition and to counteract any loss of adhesion properties occasioned by the introduction of, the wax.)

While for the record skin I3 a metal foil, for example of unannealed or annealed aluminum or of lead. may be fairly satisfactorily employed, we

Products Co., Inc., F 2d '75, certiorari denied- November 23, 1936) as ordinarily signifying simply such materials; cellulose esters (e. g.,

acetate or nitrate) caused to assume sheet form either before incorporation in the record or as an incident of that incorporation (e. g., the record or body material being dipped into a solution of such esters and thereafter exposed to evapora tion) chlorinated rubber in sheet form; and the like.

'A particularly favorable combination of skin and body material in a record for direct vertical impression recordation has proven to be any one of the organic n-terials mentioned in the preceding paragraphs, in thickness of the order of a major fraction of one mil (thousandth of an inch), with an asphalt composition (desirably containing the small percentages of wax and resin above mentioned) having a hardness of a general order indicated by the penetration figures (a) 5 mm., (b) 10 mm., and (c)-30 mm.--penetrations respectively obtained under test with a standard A. S. T. M. penetrometer with (a) 150 grams at 40 F. in 60 seconds, (b) grams at 77 F. in 5 seconds, and (c) 50 grams at F.

. in 5 seconds, which figures we believe to indicate an order of hardness appreciably below that considered satisfactory for the body material of compound records intended for any direct lateral recordation.

In respect of transverse compression and longitudinal extension-which are the strains imparted to the skin in recordationthe skin materials mentioned are characterized by elasticity. In respect of the longitudinal extension the limit of this elasticity is relatively low and is readily exceeded to leave a permanent impression in the record; but even when this limit is exceeded there always occurs a material restoration of the skin material. Very roughly it may be considered that the temporary deformation, measured in terms of displacement of the point or region to which force is applied, is proportional to the applied force; that this temporary displacement will be reduced upon withdrawal of the force, by a figure representing restoration; and that this restoration figure is very roughly constant (excepting that at displacements involving only a small overstepping of the elastic limit it tends to reduce somewhat, and at displacements involving no overstepping of the elastic limit the restoration will simply equal the temporary displacement). These tendencies, observed under equivalent tests ofthe skin materials alone, are essentially characteristic also of the combination of skin with body materials such as we have set forth as preferred; but of course the absolute amplitudes of the restorations (as well as of original displacements for given applied impulse) are very much reduced by the adhering boely material. This limitation on the strong restoration tendencies of the preferred skin materials is dependent on the inertness of the body material, for which reason among others, the inherently inert pitch (e. g., asphalt) forms an especially favorable body material.

Some of the effect of these tendencies may be observed in Figure 4, which comprises a typical approximate curve of permanent deformation or groove depth caused by operation of a record ing tool of the nature above described on a record of the preferred type abovemcntioned, plotted against record-ward biasing force. From this curve it will be seen that a groove depth of onehalf of a mil may be obtained with a biasing force of some 33 grams; that by modulating such a force by 9 grams peak (i. e., between 42 and 24 grams) depth from about 1 mil to .1.mil (e. g., +.5 mil, --.4 mil) may be obtained; that a reduction of the force to half (i. e., to 16 grams) reduces the depth to only some .02 mil; and that a reduction of the force to about a quarter of the original 33 grams eliminates the grooving altogether, being about the elastic limitin other words, a threshold of roughly 8 grams is provided by the elasticity of the skin material.

this example the mean biasing force'is only four sensible effect thereon. Of course, recordation may be effected with a mean biasing force of say 42 grams (for example modulated by 18 grams peak to result again in substantially full modulavthat the ratio of that force to the threshold, or

maximum force not exerting a significant effect, will not be more than ten. In this way 'we' obtain the latitudev to employ a reasonable biasing force in reproduction (say, in the example given, some 5 grams) with assurance that the record will be negligibly impaired, and at the same time avoid the necessity for the high biasing and modulating forces in recordation which, would be required were good record life to be secured by the alternative procedure of making the record much less impressible. And it is to be noted that the peak modulating force required for substantially full modulationis much less than the mean biasing force, and even than that force minus the threshold force.

It is particularly to be noted that oscillations of low amplitude (which among others the higher frequency oscillations almost always tend to be) are not seriously discriminated against, because the oscillatory forces which represent them are- 1 superimposed on (algebraically addedto) the steady mean biasing force; while these oscillatory forces may themselves be altogether below the threshold force value, they are only applied to the record as variations of a force which is of course much higher than the threshold, and the latter therefore does not militate againstthem. This is an important incident of the vertical recordation.

It is true that the threshold force value has tended to be materially lower and the preferred ratios exceeded in the case of records having a skin of metal foil or the like, other things being equivalent; but wedo not intend a disclaimer of the use of such skin materials except as it may be expressly or necessarily implied in any particular one of our several appended claims.

In Figure 5 there is a schematic showing of a reproducer I5 operating on the record I of earlier figures, which is supportedfor example on the turntable 9; the main portion [5a of the reproducer may be for example held slightly spaced from the record, but may carry the pivoted arm I5b to which the stylus I6 is attached,

and a spring I50 may urge the stylus against the record with a normal force for example of 5 grams (and serve to couple the arm l5b to the responsive element I5d internal of the reproducer). The stylus "5 may for example be of substantially the same size, shape and disposition as the recording tool 3, insuring most perfect following of the groove ,undulations. The record-ward biasing force on the stylus is of course subject to some variation as the stylus rises and falls in travelling along the groove, but will rarely (and then to minute extents only) rise above the typical record threshold value of some 8 grams. ,With a recording tool and record as above' described it has been foundpossible to, record with between and 200 grooves per inch (measured transversely to the grooves) at a mean displacement remain more diifuse and less wholly composed of lateral displacement variations than in otherwise analogous lateral recordation.

While we intend no unnecessary limitation thereto, we have found it desirable to arrange the system so that at the different sound frequencies to be recorded the peak forces applied to the record, and the peak modulations of the groove depth, are intermediate between constant velocity and constant amplitude, or tend to be of one of these characters over a certain frequency range or ranges and of the other over other ranges. for example may yield a record wherein the peak high frequency modulations are somewhat smaller than the peak low frequency oscillationsbut by no means smaller to theextent of inverse proportionality to frequency, as would result from the use of a simple constant velocity basis.' This practice has been schematicallyindicated in Figure 2 by the filter I lshown in the leads 4a to the-recorder, although it will be understood that a large part of this adjustment may be obtained automatically by an apt design of the recorder proper (including the dimensioning of the piezo-electric bending unit 4) In reproduction a converse practise is of course followed to obtain proper balance between different frequency components in the we have illustrated another as 3a in Figure 6.

This consists of a simple cylinder of sapphire, for example of .005." to .007"; diameter, with a plane end 3b to which the cylinder axis is at right angles. ness of the record skin and to allied considerations, it is usually somewhat to break or round the edge 30 of the end 3b; but we have preferred always to keep the degree or radius of such a rounding at as low a value as dependably precludes rupturing of the particular record skin ma-v terial with which it is to be employedtypically with the organic materials abovementioned at a small fraction of one thousandth of an inch. A typical disposition of the tool is with its axis lying at least approximately ina plane which is normal to. the record and which includes either the groove being impressed by the tool or a tangent to that groove at the point of tool engagement; with its axis forming an angle of the order of 45 with the groove or -groove tangent;

and withits end 3a preferably facing against,

rather than with, the record movement. It will of course be understood that the reproducing stylus for recordings made with such a tool is advantageously similarly formed and disposed,

- except that a somewhat greater radius of rounding of the edge may sometimes be desirable.

. The record of earlier figures has beenillustrated as a single-faced one; we have found, however, that advantages of ease of manufacture and complete freedom from warping tendencies, as well as compactness and economy, result from According to the nature and thickthe formation of the records as double-faced ones. Such a record has been illustrated as 20 in Figure 7, wherein the base appears as the central core 2| of thin metal or cardboard, the body material in two layers 22 on both faces of the core 2|, and the skin material as 23 on the two outer body material surfaces. Such a record is conveniently formed, for example, by drawing the core material in a continuous strip through a heated bath of the body material to acquire a coating of the desired order of thickness on both faces; and either rolling against both body material surfaces (while hot) respective sheets of the skin material, or passing the coated core through a solution which upon drying thereon provides the skin. Discs may be stamped or otherwise cut from the continuous record strip soformed, and if desired the edgesfrom one skin to the other-may be treated with any solution which upon drying will complete the enclosure of the body material by the skins. The symmetry of such arecord about the core obviates warping by reason of temperature changes and like causes. Because of this obviation the base or core 2| may be made thinner and itself less strong than otherwise practical,

thereby rendering the record, in spite of its double-faced nature, very thin. And there is of course obvious the compactness and economy of tworecording surfaces in a bulk of little greater order than otherwise required for one.

From the record of Figure '7 there may be omitted one of the layers 22 of body material, to result in a single faced record which is, however, better proof against warping than a simple singleface record with similar base 2|. Again, the base or core 2| may be omitted from the record of Figure 7, so that the two body material layers merge together; in such a double-faced record each skin and the immediately adjacent body material may be considered as the base for the other skin.

While it is desirable, as pointed outv above, to render the record stiff enough to lie closely against the turntable or other record support (so that the forces imparted through the tool shall be substantially fully effective in forming and modulating the groove, and not materially said skin.

dissipated in temporary flexing of a region of the record), record flexibility may be tolerated if that record portion which is under the tool from moment to moment is kept in contact with the supportwithout reliance on any more than a very small portion of the downward pressure of the tool. It is sometimes advantageous to provide discrete means for insuring this contact or bottoming, in the form of one or more rollers or brushes biased against the record immediately in front of and/or behind the tool; accordingly in Figure 2 we have shown a brush 26 immediately in front of the recording tool 3, the brush being carried for example by a spring 21 secured to the bracket B and by that spring biased against the record. The brush 26 may-for example be made of tough felt, and will incidentally serve to clean the record surface prior to recordation thereon. Of course a roller may be employed instead of a brush, and either may be employed either in front -of and/or behind the stylus.

While we have described our inventionin terms of typical embodiments thereof, we do not intend any limitation thereof by the details of those embodiments, as it will be obvious that these may be Widely varied without departure from the spirit of the invention; rather we intend in the following claims to set forth our invention as broadly as the stateof the art will permit.

We claim:

1. In a recording system having a recording tool and-means for effecting a bias of the same against a record: the combination of sound-responsive means for modulating said bias; and a record for traversal by said tool, comprising a thin skin of chlorinated rubber and therebeneath a substantially inert, yieldable body material.

2. In a recording system having a recording tool and means for maintaining the same in contact with a relatively moving record: the combination of sound-responsive means for vibrating said tool, and a record for traversal by said tool, comprising a thin skin of chlorinated rubber and therebeneath a substantially inert, yieldable body material of at least a few times the thickness of ORVILLE M. DUNNING. CHARLES 'r. JACOBS.-

Images