The method described has provided consistent data as shown by the data analysis in Table 5. The coefficients of variation for all of the absorbance measurements ranged from 0.6% - 3.3% with a mean value of 1.28%. The colour reaction upon which this test depends is time and temperature sensitive and consistency in these parameters should be strict throughout. Once fixed the colour appears to be stable for approximately 10 minutes at room temperature indicating that colorimetric readings should be taken immediately after the sample is removed from the water bath. The preparation of control samples has been described by a number of workers and although the method used here is a convenient approach, account should be taken of the moisture loss associated with the use of the microwave oven (~20%). Clearly the solute-solvent interaction that causes the loss of linearity in the absorbance/concentration data is a significant problem and one which could result in conversion errors. Some of the higher absorbance readings obtained in this experiment were approaching the upper usable limit of the calibration curve and sample concentrations should be adjusted so that they are accommodated by the linear section of the graph. Although this method of analysis is primarily developed to measure the pungency of onions, its value in providing a comparative measure for garlic clones is evident. The major limitation of using pyruvic acid determination is its lack of specificity, i.e. it is an indirect measure of total thiosulphinate and cannot provide any differentiation of the individual alk(en)yl thiosulphinates present in disrupted garlic tissue. Although this method lacks the specificity of HPLC it has been developed further by a number of workers , and while the modified methods provide greater linearity and stability they are somewhat cumbersome and time-consuming. The method described by Jager measures the DNPH of pyruvic acid after differential extraction to remove unreacted hydrazine derivative and non-pyruvate hydrazones. Calibration is afforded by the use of highly purified, recrystallised DNPH of pyruvic acid dissolved in ammonia.

Similarly, Kornberg describes a method based on the oxidation of reduced diphosphopyridine nucleotide (DPNH) by pyruvate in the presence of excess lactic dehydrogenase although a stoichiometric relation between standard pyruvate added and DPNH oxidised was obtained only when freshly prepared solutions of DPNH were used. Adaptation of the current method has avoided the complexities associated with these other methods and has, within known limits, provided consistent data. Over the past fifty years workers have adopted a number of diverse methods in attempts to quantify the pyruvate and thiosulphinate levels in various Allium species. A summary of the methods used and the quantities of pyruvate/thiosulphinate detected are shown in Table 6. The original method of Jäger is in excellent agreement with current C18- and Si-HPLC results although the high value reported by Miething is possibly due to decomposition of the allicin standard in ether. The results of this study are in good general agreement with other workers irrespective of the methodology employed.

Determination	Literature	Method	Concentration ( μM/g fresh weight)
			Pyruvic Acid	Thiosulphinate
Pyruvic acid Pyruvic acid Thiosulphinates Allyl mercaptan Diallyl sulphide Allicin Allicin Thiosulphinates	Present (12, 13) (17) (18) (19) (20) (21) (22)	DNPH derivative of pyruvate DNPH derivative of pyruvate Si-HPLC GC, headspace GC C18-HPLC C18-HPLC Si-HPLC	35 - 60 47 - 63 - - - - - -	(17 - 30)   14 - 36 15 - 30 12 - 22 22 - 30 18 - 37 26 - 40

Table 6. Comparison of results of pyruvate / thiosulphinate content of garlic clove homogenates

Statistical Analysis

Having determined the thiosulphinate (pyruvate) values for each of the samples clones, the results were filtered according to isozyme group and the data analysed to see if the thiosulphinate values of isozyme groups were significantly different. An analysis of variance test (ANOVA) was applied to the data as shown in Table 7 and Table 8. The analysis indicates that isozyme group is unlikely to have any effect on total thiosulphinate value when clones from those groups are grown under similar environmental and cultural conditions to those used in this study.

Garlic Clone	Isozyme Group
	1	2	4
California Late Carpathian Cristo Dominics French Red Freudenberger Georgia Fire German Red Inchelium Israeli Leningrad Morado Music Red Janice Red Rezan Roja Russian Redstreak Yerina Yugoslavian Count Sum Mean Variance	18.05 28.78 25.41 32.70 27.53 34.37 32.13 31.82 8 230.8 28.85 27.87	35.43 33.41 32.67 31.07 4 132.6 33.15 3.27	33.96 37.20 30.97 17.58 35.22 18.19 27.49 7 200.6 28.66 63.86

Source of Variation	Sum of Squares	Degrees of Freedom	Mean Squares	F	P-value
Between groups Within groups Total	60.8 588.0 648.9	2 16 18	30.43 36.75	0.828	0.455