WHAT IS A GENE? 173 



galactose has been analyzed by means of transduction heterogenotes 

 produced by the use of phage X. One affects the enzyme — galactoki- 

 nase, which phosphorylates galactose. A second, divisible into at least 

 four mutable sites, concerns the galactose phosphate-uridyl transferase 

 which incorporates galactose phosphate in a uridine nucleotide. This is 

 the same enzyme that is absent in cases of human galactosemia. 



Clearly genetic techniques have made possible an astonishingly de- 

 tailed analysis of the structure of the DNA molecule. It is far from exact 

 and far from complete, but it exceeds anything yet attempted by chemi- 

 cal techniques, where the first formidable problem is the separation of 

 the thousands of species of molecules that make up a DNA preparation. 

 When such a high order of specificity exists, chemistry needs to be sup- 

 plemented by biological techniques; in fact, genetics is creating a novel 

 chemistry of its own. It could be described as splitting the gene. The 

 high power of resolution of genetics can be indicated by the facts in 

 Table 6.4. 



In view of the estimates and assumptions that go into such calcula- 

 tions, judgment of the exactness of these figures should be suspended. 

 So far, not e\en in phages can a reasonable estimate be made of the 

 amount of DNA that is genetic. Furthermore, in the mouse and in corn, 

 the minimum map distances measured would certainly be vastly exceeded 

 if more offspring could be studied. It is the smallest distances which 

 reveal the resolving power of genetics, and from them we can see that 

 units as small as one millionth of the linkage map can be identified, 

 measured, and, to some degree, characterized. Such units are at the 

 level of individual nucleotides. This raises the following questions: 

 What is the unit of replication? How large is the smallest sequence of 

 nucleotides whose pre-existence is needed for its own duplication? In 

 duplication of a single gene, the nucleotides do not arrange themselves 

 at random into any possible sequence, but only into the one existing 

 in the template. How many nucleotides are needed for a unit to take on 

 this function? Clearly one pair of nucleotides does not determine its 

 own formation even though what seems to be a difference of one pair is 

 replicated in the DNA molecule. 



BIBLIOGRAPHY 



Beadle, G. W., 1955. Gene structure and gene function — L. J. Stadler Memo- 

 rial Lecture. University of Missouri, College of Agriculture, Agricultural 

 Experiment Station, Research Bulletin 588. 67 pp. (Includes L. J. Stadler's 

 last paper, 1954. The gene. Science, }20:8ll-819.) 



