On this page
Every NEET aspirant learns that in double-stranded DNA, A = T and G = C. It is easy to forget that these are measurements, not definitions, and that someone made them before anyone knew DNA was a double helix. That person was Erwin Chargaff. This post covers how he measured the bases, works through his real data, and shows the mistakes students make with his rules.
Who was Erwin Chargaff?
His National Academy of Sciences biographical memoir, written by his first student Seymour Cohen with Robert Lehman, records that Chargaff was born on 11 August 1905 and died on 20 June 2002. His family moved from Czernowitz to Vienna, where he studied chemistry and earned a doctorate in 1928. He held a postdoctoral fellowship at Yale, then worked at the University of Berlin until 1933 and briefly at the Institut Pasteur in Paris. In 1935 he joined the Department of Biochemistry at Columbia University, where he was a faculty member until 1982.
For his first decade at Columbia he studied lipids and blood clotting. He was trained as an analytical chemist, and the memoir notes that he had never imagined he would help solve a major problem in cell biology.
What made him switch to DNA?
In 1944 Oswald Avery and his colleagues reported that DNA carried the hereditary change in pneumococcus. The memoir says that "almost alone among the scientists of the time", Chargaff accepted the report and concluded that genetic differences among DNAs must show up as chemical differences. He was the first biochemist to reorganise his laboratory to test this. Our post on Avery, MacLeod and McCarty explains the experiment that convinced him.
How did he measure the bases?
His laboratory used partition chromatography on paper and ultraviolet spectrophotometry. They broke DNA down into its bases, separated the four on paper and estimated each one from its ultraviolet absorption. According to the memoir, they routinely recovered 96 to 98 percent of the bases, to make sure no unknown base was hiding.
The results overturned an older claim by P. A. Levene that DNA contains the four bases in equal amounts. By 1949, the memoir says, Chargaff could state that in all DNAs tested the molar ratios of purines to pyrimidines, adenine to thymine, and guanine to cytosine approached 1.0. The ratio of A+T to G+C, however, differed between organisms.
Worked example: Chargaff's salmon sperm data
In The composition of the desoxyribonucleic acid of salmon sperm (1951), Chargaff, Lipshitz and Green reported the average amount of each base, in moles per mole of phosphorus: adenine 0.280, guanine 0.196, cytosine 0.192 and thymine 0.274.
A to T: 0.280 / 0.274 = 1.02.
G to C: 0.196 / 0.192 = 1.02.
Purines to pyrimidines: (0.280 + 0.196) / (0.192 + 0.274) = 0.476 / 0.466 = 1.02.
A+T to G+C: (0.280 + 0.274) / (0.196 + 0.192) = 0.554 / 0.388 = 1.43.
As percentages: the four add to 0.942, so A = 29.7%, T = 29.1%, G = 20.8% and C = 20.4%.
The paper describes salmon DNA as an "AT type", with A and T outweighing G and C by about 40 per cent. All the ratios match the published table.
Now an exam-style version. A double-stranded DNA sample has 20% adenine. What are the other bases?
T = A = 20%.
G + C = 100 - 40 = 60%, so G = C = 30%.
Common mistakes:
Applying the rules to single-stranded nucleic acids. They hold for double-stranded DNA. The memoir notes Chargaff's RNA samples had very different compositions from DNA. A single-stranded viral DNA need not follow them either.
Assuming A+T equals G+C. It does not. In salmon DNA the ratio is 1.43. Only A+G = C+T follows directly from pairing.
Reading the rules as a law of nature, not a measurement. His ratios were 1.02, not exactly 1, because measurements have error.
Why didn't Chargaff find the double helix?
He saw the pattern but did not explain it. In the 1951 paper he asked whether the regularity was accidental or "an expression of certain structural principles", and answered: "It is believed that the time has not yet come to attempt an answer." The memoir adds that the pairing meaning was not stated explicitly by Chargaff.
Two years later, Watson and Crick's 1953 paper proposed a structure in which A pairs with T and G with C, and noted that it had "been found experimentally that the ratio of the amounts of adenine to thymine, and the ratio of guanine to cytosine, are always very close to unity", citing Chargaff's laboratory. The structure also drew on X-ray work described in our profile of Rosalind Franklin and Photo 51. The NCERT Class 12 chapter credits Chargaff's observation as one basis of the Watson-Crick model.
Chargaff felt his work had not been properly credited. He still received many honours, including the National Medal of Science in 1974, "for fundamental chemical and biological studies establishing the basis for modern concepts of the mechanisms of protein synthesis and the genetic role of nucleic acids."
What can a student take from his way of working?
Take a surprising result seriously. He changed his whole research programme after reading one paper.
Check your recovery. Measuring 96 to 98 percent of the bases told him his method was not missing anything.
Report the numbers, not just the conclusion. His published ratios still let you recheck his work.
Notice what your data cannot yet explain. He stopped short of a structure, and others completed the step.
Practising base-ratio problems
Questions on Chargaff's rules are quick marks if you set them up correctly. The Fermi Problem Library has worked problems with a step-by-step solver that explains the reasoning rather than printing an answer, which helps when you are unsure whether a strand is single or double.
Try it now: a double-stranded DNA has 18% cytosine. Work out the percentage of each base, then check that A+G equals C+T.
