U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter 4

Table of Contents


Development and Description of Electrical

Accounting Machines

Fortunately, the inventor of the electrical tabulating equipment, Dr. Herman Hollerith, left a very clear description of the machines, with several photographic illustrations.1 Dr. Hollerith who was a statistician with a doctor of philosophy degree from Columbia College was employed as a special agent by the Census Office in 1880.2 There, he came in contact with Col. John S. Billings who was consultant to the Census Office in addition to being librarian of the medical library of the Surgeon General's Office. Garrisons3 says: "From this time in 1878 to 1912, Billings was active and prominent in the supervision of the vital statistics of the 10th, 11th, and 12th Census of the United States, 1880, 1890, and 1900." Colonel Garrison adds that the Compton Reviewer in the London Times, 22 July 1915, states: "Before 1880, when Dr. Billings took charge of the Vital Statistics of the U.S. Census, they were worse than useless. For three decades, the 10th, 11th, and 12th census, Dr. Billings was a volunteer worker in this field of statistical inquiry and from a state of chaos he brought the vital statistics of the United States to their present satisfactory condition."

It seems clear that Dr. Billings discussed with Dr. Hollerith

1Hollerith, H.: An Electrical Tabulating System. Columbia College, New York City School of Mines, Quarterly, A Journal of Applied Science. 10: 238-255, October 1888 to July 1889.

2The Punched Card, 1952-53, vol. I. Detroit: Punched Card Publishing Co.

3(1) Garrison, Fielding H: The Scientific Work of John Shaw Billings, in Biographical Memoirs. Washington: National Academy of Sciences, 8: 385-416, August 1917. (2) Proceedings, American Association for the Advancement of Science, 1891, 40: 407-409, published by the Permanent Secretary, Salem, Mass., August 1892.


the possibility of a machine such as Dr. Hollerith eventually developed. Garrison4 states that, in 1880, Dr. Billings suggested that the various statistical data of the living and the decedent might be recorded on a single card or slip by punching small holes in it and that these cards might then be assorted and counted by mechanical means according to any selected group of these perforations. Garrison adds that this suggestion was taken up and applied by Dr. Hollerith in electrical counting and integrating machines first used by the Census Office in 1890.

Dr. Raymond Pearl5 advances the opinion that it was Dr. Billings and not Dr. Hollerith who was entitled to the credit for the development of the Hollerith machines. According to information obtained from an article by Dr. Billings, apparently the same referred to by Colonel Garrison, Dr. Pearl states:

That the system collected by the census for each living person or in systems of death registration for each decedent might be recorded on a single card or slip by punching small holes in different parts of it, and that these cards might then be sorted and counted by mechanical means according to any selected grouping of these perforations was first suggested by Dr. Billings in 1880. This suggestion was taken up by Mr. Herman Hollerith, and by him had been elaborated into practical shape the system which is now used in the compilation of the Vital Statistics Division of the 11th U. S. Census and which had been adopted for the compilation work of the recent Austrian and Canadian census.

Dr. Pearl also quotes from an article by Dr. Hollerith:6 "While engaged in the 10th census, that of 1880, my attention was called by Dr. Billings to the need of some mechanical device for facilitating the compilation of population and similar statistics." Hollerith adds that Billings then enumerated a number of reasons that made such a system necessary. In conclusion, Hollerith states: "These were the considerations which prompted me to take up this problem, the result of which researches after years of experimental work are embodied in the apparatus or system which I will now briefly describe."

Before detailing Hollerith's description of his machine, it is

4Garrison, Fielding H.: John Shaw Billings; A Memoir. New York: G. P. Putnam Sons, 1915.

5Pearl, R.: Some Notes on the Contribution of Dr. john Shaw Billings to the Development of Vital Statistics. Bull. Inst. Hist. Med. 6 (5): 387-393, May 1938.

6Hollerith, Herman: The Electrical Tabulating Machine. London Royal Statistical Society. 57: 674-682, December 1894.


interesting to review the background of the use of the perforated card. The principle of using a perforated card was first applied to weaving machinery. In 1780, Joseph Marie Jacquard invented a machine in which perforated cards controlled the pattern of hooks and needles which were lifted or dropped in order to pass a woof over or under the warp. One perforated card was required for each such cross thread in the pattern, and these cards were laced into a chain.

Later, Charles Babbage, a distinguished mathematician in England, obtained a governmental grant to build up a computing machine utilizing punchcards, but during the course of its construction he became interested in modifying it into an analytical engine. The Government refused to provide the necessary additional funds, but the machine was eventually finished about 1840. A Swedish printer named Scheutz developed further refinements in the mechanical operation of this machine, in which punchcards replaced finger operation. No information was found that such a machine was used in statistical work. In fact, no reference has been found by the author in the writings of Dr. Hollerith or of Dr. Billings of their consideration or even knowledge of this perforated card system as used in the weaving industry. It appears to be the opinion of some writers on the subject, however, that it was upon the sequence of inventions and improvements stemming from the abacus, Pascal calculator, tally sticks, and Jacquard loom control card that Dr. Hollerith based his idea of using the perforated card for statistical purposes.

It appears that Dr. Hollerith first used his equipment in connection with the 1886 mortality records of the city of Baltimore. He so states in his patent application7 and in a personal communication.8 He does not state, however, where the work was done nor whether it was completed.

Hollerith's daughter, Miss Virginia Hollerith, states that she has seen a reference in his correspondence to the equipment's being used in New Jersey before its use in the Surgeon General's Office in 1889.

Dr. Hollerith first used a tape for recording data by perforations or punches. The tape was soon replaced by a card similar

7Patent application 4 Jan. 1887. Patent granted (No. 395781) 8 Jan. 1889. Prior applications 23 Sept. 1884, 27 Oct. 1885, and 27 Sept. 1888.

8Personal communication, Herman Hollerith to Mr. Wilson, 7 Aug. 1919.


to but larger than the one used in 1886 for the statistical work with the health department of the city of Baltimore and the one used in 1889 in the Surgeon General's Office. Before the adoption or patenting of the pantotype punch (fig. 1), he used a conductor's punch and hence could perforate only the top, bottom, and ends of the card. The punch shown for use in Baltimore on his patent application of 4 January 1886 had three rows at the top of the card and three at the bottom for perforating. There were 32 spaces provided in each row. The middle of the card was used for writing data.

As has been stated, it is fortunate that Dr. Hollerith left a clear and lucid description of the development of his tabulating system. In an article printed in 1889,9 Dr. Hollerith discussed in some detail and at length the necessity for such a machine. He pointed out how impossible it was to compile for a constantly increasing population the information that was required for the United States census. Much of the material on the forms as returned by the enumerators in the districts was unavailable for publication, since it could not be assembled in statistical tables. After discussion of the necessity for the development of such a system of electrical tabulating machines, Dr. Hollerith stated:

In a census the enumerator's districts form the statistical units of the area, and a suitable combination is arranged to designate each such district. A card is punched with the corresponding combination for each person in such enumeration districts, and the cards of each district are then numbered consecutively, in a suitable numbering machine, to correspond with the number assigned to the individual records in the enumerator's returns. This combination of holes and this number will serve to identify any card. Should any card become displaced, it is readily detected among a number of cards by the fact that one or more of these holes will not correspond with the hole in the balance of the cards. By means of a suitable wire or needle, a stack of a thousand or more cards can be tested in a few seconds, and any misplaced card detected.

* * * as a combination of these holes usually designating the enumerators' districts are the same for all cards of that district a special machine is arranged for punching these holes. This machine is provided with a number of interchangeable punches which are placed according to the combination it is desired to punch. Five or six cards are then placed in the punch against suitable stops, and by means of a lever the corresponding holes are punched through these cards at one operation.

9See footnote 1, p. 36.


Then with reference to the perforation of the individual records, he described what is known as a pantotype punch, which he invented (fig. 1).

He says:

Individual records are now transcribed to the corresponding cards by punching according to a pre-arranged scheme as described above. For this purpose, what may be known as a keyboard punch is arranged in which the card is held in a frame, while the punch is moved over the card in any direction by means of a projecting lever provided with a suitable knob or handle. Below the knob is a keyboard provided with holes lettered and numbered according to the diagram of the card, and so arranged that when a pin projecting below the knob is over the hole the punch is over, the corresponding space on the card is punched. With such a keyboard punch, it is of course apparent that a perfectly blank card may be used, one corner, however, being cut off to properly locate the card in subsequent operation.

The following paragraph is of special interest to the Army Medical Service:10

Heretofore, reference has only been made to the compilation of a census, but these methods are equally applicable to many other forms of

10Correspondence, Office of the Surgeon General, U.S. Army, in old Records Division of Adjutant General's Office, Archives Building, Washington, D.C. The authors wish to express their appreciation to Dr. Clifford J. Maloney for calling their attention to this correspondence and to Miss Mable Deutrich for making it available to them.

    1. December 1886. Capt. Fred C. Ainsworth reported to the Surgeon General's Office and was placed in charge of all medical records (including those for the Civil War period) and medical statistics.

    2. September 3, 1888. A contract was signed by Herman Hollerith for his equipment to be installed in the Surgeon General's Office.

    3. January 9, 1889. Herman Hollerith, by letter, informed the War Department that equipment was installed in the Surgeon General's Office.

    4. April 8, 1889. Herman Hollerith requested a certificate of efficiency and usefulness of his equipment to be used for display with his equipment at the exposition, Paris, France.

        (1) Capt. Fred C. Ainsworth, in his endorsement, stated that, although the equipment appeared to be satisfactory, additional trial was necessary to determine its full usefulness. (2) Maj. Charles R. Greenleaf, for the Surgeon General, stated that such certification was contrary to War Department policy.

    5. July 3, 1889. A War Department directive was issued consolidating Civil War Volunteer Medical Records (SGO) and Muster Rolls (AGO), in a new and independent Record and Pension Division with Capt. Fred C. Ainsworth in charge.

    6. July 9, 1889. Captain Ainsworth recommended renewal of Hollerith contract for the Surgeon Generals Office for the fiscal year ending June 30, 1890. Captain Ainsworth stated that 50,000 cards had been prepared but equipment was needed for sorting and tabulating them.

    7. July 16, 1889. A final order was issued organizing the new Record and Pension Division, Captain Ainsworth was relieved in the Surgeon General's Office by Maj. Charles Smart, MC, in whose division the machines remained.

    8. July 22, 1889. A contract for the continued use of the machine in the Surgeon General's Office was signed with Herman Hollerith for the year ending June 30, 1890. No record has been found relative to the termination of this contract.


FIGURE 1.-Pantotype punch


statistical compilation, as for example the various forms of vital statistics. Figure 2, for example, represents the diagram of the card as at present used in the Office of the Surgeon General, U.S.A., for compiling the Army health statistics. The data relating to the month, the post, the division, and the region to which the record relates is recorded by punching a hole in each of the divisions across the end of the card by means of the machine with interchangeable punches as before described. This portion of the record corresponds almost exactly to the record for enumeration numeration districts of the census. The individual record is then transcribed to the card by punching in the remaining spaces with a keyboard punch as before described. Such a card allows a complete record, including the following data for each individual; rank, arm of service, age, race, nationality, length of service, length of residence at the particular post, whether the disease was contracted in line of duty or not, whether

FIGURE 2.-Card used in the Surgeon General's Office, 1889


admitted to sick report during the month or during the previous month, the source of admission, the disposition of the case, or whether remaining under treatment, the place of treatment, the disease or injury for which treated, and finally the number of days treated.

*      *     *     *     *


Between 40 and 50 thousand such records are received annually, and from these are compiled the various health statistics pertaining to our Army.

Dr. Hollerith then adds a very important paragraph:

A card has just been arranged for the Board of Health of the City of New York to be used in compiling the mortality statistics of that city. The record for each death occurring in the City of New York, as obtained from physicians certificate, is transcribed to such a card by punching as before described.

The Hollerith equipment (fig. 3), in addition to the punches, consisted of press, counters, and sorting box. The assembly of the three is shown in figure 3D.

The press or circuit-closing device consisted of two parts: A reciprocating box with a number of projecting spring-actuated points for each separate space on the card used and a hard rubber plate placed below the box with a hole or cup for each space on the card. Each such cup was partially filled with mercury that contacted, through the hole in the bottom of the cup, with a wire leading to a counter or a sorting box. For operation, the card was placed on the rubber plate. The pin box of the press was then depressed by the handle. The pins passed through any perforation in the card and made contact with the mercury in the corresponding cup. Thus, contact was made with an electrically charged wire below leading to the counter or to the sorter,

Dr. Hollerith described the counters as follows:

A number of mechanical counters are arranged in a suitable frame as shown in Fig. 3B. The face of each counter is three inches square and is provided with a dial divided into 100 parts and two hands, one counting units, the other hundreds. The counter consists essentially of an electromagnet, the armature of which is so arranged that each time it is attracted by the closing of the circuit it registers 1. A suitable carrying device is arranged so that with each complete revolution of the unit hand the hundred hand registers 1. Each counter thus registers or counts to one hundred hundred or 10,000 which will be found sufficient for ordinary statistical purposes. The counters are so arranged that they can readily be reset at 0 and all are removable and interchangeable, the mere


FIGURE. 3-The Hollerith tabulating system. A. Press. B. Counters. C. Diagram of sorting box. D. Assembled unit

placing of the counter in position in the frame making the necessary electrical connection with the magnet.

For the purpose of sorting the cards according to any group of statistical items or combination of two or more such items, the sorting box shown in Fig. 3C is used. This consists of a box suitably divided into compartments, each one of which is closed by a lid. Each lid L, as shown in Fig. 3C, is held closed against the tension of a spring S, by the catch a, in the armature A. If the circuit is closed, through the magnet E, the armature A is attracted, thus releasing the lid L which is opened by the spring and remains open until again closed by hand.

As the cards (Census) are punched, they are arranged by enumerators' districts which form one unit of area. The first compilation that would be desired would be to obtain the statistics for each enumeration district according to some few condensed grouping of facts. Thus, it might be desired to know the number of males or females, of native born and foreign born, of white and of colored, of single, married and widowed, the number of each of center groups of ages etc., in each enumeration district. In order to obtain such statistics, the corresponding binding posts on the back of the press frame are connected by means of a suitable piece of


covered wire with the binding posts of the counter upon which it is decided to register the corresponding facts. A proper battery being arranged in circuit, it is apparent that if a card is placed on the hard rubber bed plate and the box of the press brought down upon the card, the pins corresponding with the punched spaces will close the circuit through the magnets of the corresponding counter which thus registers one each. If the counters are first set at 0 and the cards of a given enumeration district are then passed through the press one by one, the number of males and of females, of white and of colored, etc., will be indicated on a corresponding counter. If it is desired to count on the counters directly, combinations of two or more items, small relays are used to control secondary circuits through the counters. If, for example, it is desired to know the number of native white males, of native white females, of foreign white males, or foreign white females, of colored males and of colored females, these being combinations of sex, race, and nativity, ordinary relays are arranged as shown in the diagram (fig. 4), the magnets of which are connected with the press as indicated. If a card punched for native white and male is placed in the press, the corresponding relays are actuated which close the secondary current through the counter magnet, native white male, thus registering 1 on the corresponding counter. If a card punched for native white and male is placed in the press, the corresponding relays are actuated which close the secondary current through the counter magnet, native white male, thus registering 1 on the corresponding counter.

By a suitable arrangement of relays any possible combination of the data recorded on a card may be counted. When it is desired to count more complicated combinations, however, special relays with multiple contact points are employed.

If it is desired to assort or distribute the cards according to any desired item or combination of items recorded in the card, it is only necessary to connect the magnets of the sorting box in exactly the same manner as has been described for the counter. When the card is thus placed in the press, one of the lids of the sorting box, according to the data recorded on the card, will open. The card is deposited in the open compartment of the sorting box and the lid closed with the right hand while at the same time the next card is placed in position in the press with the left hand.

It is of course apparent that any number of items or combination of items can be counted. The number of such items or combinations which can be counted at any one time being limited only by the number of counters, while at the same time cards are sorted according to any desired set of statistical facts required.

It seems to be apparent that Dr. Hollerith's ingenious machine, at least in the beginning, was more a selective counting rather than a sorting device. Hollerith comments, in referring to the installation in the Surgeon General's Office, about


FIGURE 4.-Diagram for combination counting

a space being provided for punching the number of days lost by each patient in the hospital or on sick report. He states in his letter to Mr. Wilson that when working with Dr. Ainsworth, later Adjutant General, he found it necessary to develop or invent a machine to accumulate (integrate) the days lost by sick soldiers. It appears that the machine was crude and unsatisfactory, since, when it was later used in New York City (July 1889), the office there discontinued the use of the entire equipment until Dr. Hollerith could develop a more satisfactory machine. This appears to have been the one used in the census work in 1890 (fig. 5). Possibly the unsatisfactory character of the integrating machine may have been the reason or at least one reason for discontinuance of the use of the equipment by the Surgeon General's Office.

In April 1889, Robert P. Porter was appointed Superintendent of the Census, then an office of the Department of Interior. On 30 July 1889, Porter appointed a commission of distinguished scientists to conduct a test of the relative efficiency and cost of the Hollerith electrical system and two manual counting


and sorting card systems. The commission was composed of Dr. John S. Billings,11 chairman, and two additional members, Mr. Henry Garnett, a geographer, and Mr. William C. Hunt, an expert in population studies, both employed as chiefs in the Census Office. Since Mr. Hunt was submitting one of the counting system in competition with Dr. Hollerith, he asked to be relieved from the commission. Mr. L. M. E. Cooke was named to replace him.

Dr. Hollerith's system has been discussed. Mr. Hunt proposed a card system, with the cards sorted and counted by hand. The information was expressed upon cards in part by the color of the ink and in part by writing upon them. A third method, submitted by Mr. Charles F. Pidgeon, also proposed a hand sorting and counting operation. In this system, the cards, called "chips," were small slips of paper of various colors, containing printed matter in various colors, these colors being used to indicate certain facts found upon the schedules. Those census facts not expressed by the colors were written on the slips.

To test these three methods, four enumeration districts in St. Louis, Mo., selected from the reenumeration of 1880, were chosen. These districts contained 10,491 inhabitants. Each contestant was required (1) to transfer the information from the schedules to his cards, (2) to tabulate this information in the form of a table, for the purpose of testing the rapidity of his method, and (3) to tabulate the information in the form of extended tables, for the purpose of testing the capacity of his method.

The testing of the three methods began late in September and early in October. On 11 November, Dr. Hollerith and Mr. Hunt having completed the test of their methods and Mr. Pidgeon having advanced far enough for the purpose, it was decided to suspend the test. The commission expressed the

11In Love's opinion, the appointment of Colonel Billings to the commission by Mr. Porter and his acceptance of the appointment showed conclusively that neither of them considered that Colonel Billings was the inventor of the Hollerith equipment or that he had any proprietary rights in it, either actual or implied. The suggestion made in 1880 by Colonel Billings, then a forceful man of 42 with a well-established reputation as an intellectual, must have made a great impression on Dr. Hollerith, then only 20 years of age. But the making of such a suggestion in no sense gave Colonel Billings any proprietary rights to Dr. Hollerith's remarkable invention. The development of the original rather crude equipment required years of brilliant and continuous effort at an early period in the practical use of electricity. Even after his initial success, more than 20 years of continued brilliant work were required to perfect it.


opinion12 that (1) the chance of error in transcribing by the three methods was approximately equal, (2) the transcribing was done by the Hollerith method in three-fourths as much time as by either of the other two, (3) the tabulation was done by the Hollerith method eight times faster than by either of the other two, and (4) estimating the population of the United States in 1890 as 65 million and the cost per day of a clerk as $2.50, it was judged that the use of the Hollerith method would result in a saving of $579,165. The report was submitted on 30 November 1889. It was approved, and Dr. Hollerith was awarded the contract.

Mr. T. C. Martin, a science writer and later a distinguished science editor, gives a most interesting graphic and comprehensive account of a visit that he made to the Census Office to study the operations of the Hollerith machines.13 Martin was much impressed by the speed and accuracy of the work done by the clerical force and by the efficiency of the electrical system. He reports that one clerk was punching an average of 1,000 cards per day after certain general information had been "gang punched" on them. He was also impressed by the system of checking for errors by visual inspection, by so-called needle sorting of common information, and by the rejection of cards during the use of the press in multiple selective counting operations.

With the use of the Hollerith equipment, only about 6 weeks were required to complete the population count as of 1 June 1890. The rough count was ready and published as early as 30 October 1890. The count of all the data, including some late reports which did not reach the Census Office until 10 November, was completed and ready for publication on 12 December 1890, the final count being 62,622,250. Since this total count and that of the various States and cities were below the figures expected, angry protests were voiced by the press and politicians. The accuracy of the count was questioned; Mr. Porter, the Superintendent of the Census, was denounced; and the ability and almost the honesty of Dr. Hollerith were questioned. But as the editor of the Electrical Engineer wrote in

12Report of a commission appointed by the Superintendent of Census, subject: Different Methods of Tabulating Census Data. Washington: Judd & Detweiler, 1889.

13Martin, T. C.: Counting a Nation by Electricity. The Electrical Engineer, 11 Nov. 1891.


an editorial that was published in the same issue as Martin's article, no count could have been more accurately determined. He adds that no doubt mistakes were made by the census enumerators in filling out the initial returns, and even in the Census Office itself, but the accuracy of the Hollerith system was far greater than that of any system formerly used. Furthermore, the cost was 40 percent below that estimated by the commission.

Mr. Martin states further that he was also impressed by the great flexibility of the system and by the multiple selective counting. By an ingenious system of electrical relays and magnets, by one handling of the card and one action of the press, population figures were able to be broken down or counted by (1) race, (2) color, (3) sex, (4) marital status, (5) nativity, (6) age, (7) State where born, and (8) certain health conditions (as insane, idiot, blind). Other factors could also be counted by the one operation. He noted that the chiefs of the various divisions of the Census Office were enthusiastic in their approval of the system, not only on account of its speed and accuracy but also on account of its great flexibility and the fields of research that might possibly be opened by it. He adds that perhaps the most practical demonstration of the value of the Hollerith system was its adoption by the Governments of Canada and Austria for their censuses. The committee on science and arts of the Franklin Institute, Philadelphia, Pa., after seeing the system in operation in Washington strongly recommended that Dr. Hollerith be granted the highest award of the Institute. Consequently, the Institute conferred on Dr. Hollerith the Elliot Cresson Medal.

Martin finally discusses an integrating machine perfected by Dr. Hollerith (fig. 5). This machine had not been included in his original equipment and was not discussed in the article described previously.14 Hollerith had found that an integrating machine was necessary to assemble days lost by soldiers. Hence he developed an integrating machine later perfected for his work in New York City. It appears that this machine (or a modification of it) was used for the census of the farm and manufacturing sections of the census. This machine (fig. 5) could accumulate totals by the use of a system of cogs, studs,

14See footnote 1, p. 36.


FIGURE 5.-Hollerith integrating machine

and electrical connections with multiple wires. Totals could be accumulated by one handling of the cards and one action of the press simultaneously with other enumerations.

Reference was also made to another change in the Hollerith equipment. Dr. Hollerith, in his original article, says with reference to the press: "An iron wire nail is securely driven through a hole in the bottom of each cup, and a wire completes the connection with the counter or sorter." Apparently the iron nail, and presumably the iron wire, were found unsatisfactory. Martin says: "The circuit is really closed first through platinum contacts at the back of the press. In this way, no difficulty is experienced from the oxidation of the mercury from the spark as would be the case without this precaution."

In conclusion, Martin adds:15

Not a little skill and judgment was necessary in perfecting the mechanical details of the Hollerith electric tabulating system that has been described

15The authors are greatly indebted to Dr. Margaret Merrell, of the School of Hygiene and Public Health of the Johns Hopkins University, for calling their attention to the report of the commission and to the article by T. C. Martin and, in addition, for the loan of the report and a reprint of the article.


above, and Mr. Hollerith freely acknowledges his indebtedness for the assistance of the manufacturers who built his apparatus. To the Pratt-Whitney Co., of Hartford, Conn., he wisely entrusted the development and construction of the keyboard punches, and Mr. George M. Bond, the well-known expert of the concern, gave the work his direct personal supervision from first to last. The electrical apparatus is the production of the Western Electric Co., whose New York representatives, Mr. H. B. Thayer and Mr. Nickel, as well as others, took a deep interest in its refinement and perfection. Of course the work itself, which will continue at the Census Office for some time yet, is watched over by Mr. Hollerith, who is under contract to the Government to furnish the apparatus and maintain it in an efficient state. It may be added that the current for doing the work is derived from the local Edison lighting circuits in Washington, being first passed through a small set of Electrical Accumulator Co.'s cells grouped in a corner of the basement, with the usual attachments for charging in series and regrouping in multiple, as well as for making tests to see that they are up to the requirements of the daily work.

In 1896, Dr. Hollerith incorporated his machine under the name of the Tabulating Machine Company,16 with a New York charter. Shortly after its incorporation, the Tabulating Machine Company purchased a tract of land in Washington on 31st Street and the C & O Canal which was later the site of the Washington plant of the International Business Machine Corporation. A small brick building on this property was used by Dr. Hollerith for several years as an experimental laboratory and workshop. This plant soon outgrew the building, and another was erected. Still later, more land had to be purchased and several other buildings were added to the plant. It was this building that Colonel Kean and Captain Love visited about 1912 to inspect the machines, when it was suggested by a Government administrative advisor that the Hollerith machines be installed again in the Surgeon General's Office. They were not adopted at that time, however, either because the rental cost was too great for the peacetime budget or because the clerical opposition was too great to be overcome.

The Tabulating Machine Company merged in 1911 with the Computing Scales Company of America and the International Time Recording Company to form the Computing-Tabulating-Recording Company. There were four machines in the Hollerith equipment used in 1911 by the new Computing-Tabulating-Recording

16Machine Methods of Accounting, Development of International Business Machine Corporation. IBMC: 1936.


Company.17 They were the mechanical key punch, a sorter, a tabulating machine for adding or integrating the totals on a card, and the lever-set gang punch which had movable levers that could be set to represent several digits so that common numbers could be punched into a pack of cards inserted into the machine. These apparently were the same machines that were used in the Surgeon General's Office in 1917 and 1918. In 1922, the printing features for the tabulating machine were announced. Love saw the first machine in the Post Office building on Pennsylvania Avenue about that time. It is interesting to note that Dr. Hollerith retired in 1914 and that Mr. Thomas Watson, whose name has been connected for many years with the IBM electric accounting machines, was elected president of the reorganized company.

It appears that the Surgeon General's Office was the first in the War Department to use the Hollerith equipment in 1917, with the possible exception of the Ordnance Department. Love understood at that time that that Department was installing equipment, although he has not been able to secure confirmation. It may have been used in one or more of the ordnance depots, even if not in Washington.

Subsequent to the patent of 8 January 1889, Dr. Hollerith had made frequent improvements in his equipment. These improvements were all covered by patents. The records in the Patent Office show at least 20 additional patents granted to him or in his name before 1920. His daughter, Hiss Hollerith, states that after the sale of his patent rights to the now International Business Machine Co., he continued as an adviser until the early 1920's.

17A Brief History of the IBM Electric Accounting Machine Division. IBMC: no date.