used. Therefore, as the Key Is studied, an understanding will be gained of the apportioning of the total 

 available punched card columns to each category of information. 



For example, eight IBM punched card columns have been committed to coding the biological 

 system (a test organism or a tumor or pathology of an organism) and this group of eight columns is 

 regarded as a single coding field , representing a major category of biology information. Considering 

 coding of test organisms only, it will be seen that test organisms might be given simple serial numbers, 

 so that the Norway rat might have been assigned Symbol 1, a species of mosquito Symbol Z, a species 

 of bacteria Symbol 3, the tobacco plant Symbol 4, etc. Had this been done, the few hundred (or 

 perhaps, few thousand) organisms that may reasonably be expected to be used in chemical-biological 

 tests might have been accommodated by considerably fewer than eight columns. Instead, the CBCC 

 recognized not only the desirability but the necessity of indicating by code the taxonomic affinities of 

 each organism used. Thus, the total field of eight columns was sub-divided into what might be thought 

 of as sub-fields. As a result, only a single IBM punched card column was assigned to coding of the 

 phyla to which organisms belong; a single column was assigned to coding of classes; a single column 

 to coding of orders; two columns to coding of families; two columns to coding of genera; and one 

 column to coding of species (or strains). As an illustration of the apportioning of IBM punched card 

 columns, this particular example has been chosen with the secondary objective of preventing any 

 initial impression that certain coding fields, such as Field E, have been assigned a number of IBM 

 columns providing a number of code symbols far beyond the needs of the CBCC coding. This misap- 

 prehension will be avoided by recognizing that, within the fields representing major categories of 

 information, the IBM columns may be committed to special sub-categories of information such as 

 phyla, classes, or orders of test organisms, the coding of each of which i£ satisfied by one or two 

 columns. 



The question as to whether the 63 available IBM punched card columns are sufficient or are 

 more than sufficient is not to be answered positively one way or the other, except in terms relative to 

 the character of the particular coding project. This is discussed in the Appendix, in terms of the 

 particular needs of the CBCC and of the Code's adaptability. 



The Detailed Nature of the Code; Its Limitations and Aspirations : 



While the CBCC functioned, there was always the reasonable assumption that any special 

 Information collection elsewhere might be advantageously indexed (coded) to be coUatable with the 

 Center's coded files. Even with the Center discontinued, it has not been unreasonable to hope that 

 the Biology Code might be accepted as, or serve as the basis for, a standard, so that information 

 might be freely exchanged between open collections and so that information of one collection might 

 be collated with that of another. 



Many of the specialized information collections now being Initiated are private and some are 

 even confidential for proprietary purposes. Thus, coding of chemical-biological information in these 

 collections may be frankly as much cryptogrammic as it is documentary. Beyond this, most projects 

 of information collection are severely subject to justifications of time and expense in coding; the 

 additional effort of coding to a standard and more complex scheme Is, at least under the present 

 situation, not apt to be viewed in many cases as offering foreseeable practical advantage. 



When an information collection is limited in scope, the variety of indexing criteria and items 

 is correspondingly less than that needed by the CBCC to index (code) information of all types. For 

 example, if the information collection were never to involve more than twenty or twenty-five organisms, 

 simple sequential code symbols of a single unit might be entirely adequate for organism code iden- 

 tification. The same would be true of categories of Information other than test organisms. Because 

 of its broad character, the CBCC Biology Code has frequently been regarded critically as too detailed 

 for practical considerations in indexing information collections of a limited nature. 



If it develops that all future collections of information about chemical-biological data are to 

 be collections of specialized nature and that the coding of each is to be referable to none of the others 

 or to no central and standard coding scheme, much of the effort spent on the CBCC Biology Code may 

 prove unfortunately to have been wasted. 



This Code is presented with the conviction that it will prove something more than a key to the 

 information collected and coded by the CBCC during its active period. It may be hoped that eventually 

 this Code or a derivative will be found universally acceptable and will serve as the basis for better 



- 9 



