such as potassium, sodium and chloride, are known to be essential both in and

Transcription

1 . JI exp. Physiol. (1969) 54, THE DISTRIBUTION OF THE PRINCIPAL INORGANIC IONS IN SEMEN FROM THE VAS DEFERENS OF THE DOMESTIC FOWL AND THE CONTENT OF CARBON DIOXIDE IN THE SEMINAL PLASMA. By M. H. EL JACK* and P. E. LAKE. From the Agricultural Research Council Poultry Research Centre, King's Buildings, West Mains Road, Edinburgh, 9. (Received for publication 1Oth January 1969) The distribution of sodium, potassium, calcium and magnesium between the spermatozoa and plasma in the semen of the vas deferens of the domestic fowl has been determined by atomic absorption spectrophotometry. Chloride in the seminal components was analysed by an electrometric titration method and carbon dioxide in the seminal plasma was analysed using gas chromatography. The contents of potassium and magnesium were much higher, and those of sodium, calcium and chloride lower in the spermatozoa than in the seminal plasma. The amount of sodium in the plasma of the vas deferens was about the same as that in blood plasma, but the amounts of potassium and magnesium were greater in the seminal plasma. The amount of chloride was small in the seminal plasma and its significance in relation to the ionic balance in the fluid is discussed. BALANCED electrolyte concentrations, particularly those of inorganic ions such as potassium, sodium and chloride, are known to be essential both in and about cells for their proper functioning. This applies to somatic and germ cells and, presumably, also mature spermatozoa but nothing was known about the ionic composition of fowl spermatozoa until recently when, coincident with this study, Quinn, White and Wirrick [1965] reported on the distribution of some cations in the semen of several mammals and the domestic fowl. Much information is available on the composition and metabolism of a wide variety of somatic cells and tissues, including knowledge of the ionic balance between cells and surrounding fluids. However it is not certain how much of this knowledge can be applied to mature spermatozoa and their functions, or whether it may help in seeking a solution to the problems of maintaining spermatozoa in vitro. Fowl semen when collected by massage is liable to be contaminated with variable amounts of transparent fluid from the cloaca [Takeda, 1959], which has been shown to be detrimental to the survival of the spermatozoa, possibly due partly to its high content of chloride [see Lake, 1966]. Fowl semen can be collected by careful massage directly from the vas deferens without contamination from other extraneous fluids [Lake, 1957] and a study has been made of the distribution of the principal inorganic ions within such semen samples using the relatively simple and sensitive technique of atomic absorption spectrophotometry. An attempt was made to study * Present address: Faculty Khartoum North, Sudan. of Agriculture, University of Khartoum, Shambat, 332

2 Composition of Fowl Semen 333 the intracellular content of ions in relation to that of the extracellular fluid (seminal plasma) in the vas deferens, in which organ the spermatozoa of the fowl apparently survive for a relatively long period [Munro, 1938]. Methods of preserving fowl spermatozoa in vitro still fall far short of the efficiency of natural storage in the vas deferens or oviducal host glands and thus a careful examination of the chemistry of spermatozoa in these organs might indicate ways in which methods for in vitro storage could be improved. Lake, Butler, McCallum and MacIntyre [1958], Takeda [1959] and Lake and El Jack [1964 a, b] have described preliminary studies of the inorganic ions and other chemical compounds in seminal plasma obtained from the vas deferens; and Hammond, Boone and Barnett [1965] and Quinn et al. [1965] reported inorganic analyses of fowl seminal plasma which was possibly contaminated with an unknown amount of transparent fluid from the cloaca. MATERIAL AND METHODS Fifty-six adult male birds of the same age and randomly selected from the breeding line of Brown Leghorns at the Poultry Research Centre, Edinburgh [Blyth, 1954], were used as semen donors. They were kept in large, single metal cages, fed a commercial breeder's ration [El Jack and Lake, 1966 a] and given 14 hr of artificial light daily. Collection of semen. - Uncontaminated semen from the vasa deferentia was collected by the especially careful application of the massage technique described by Lake [1957]. It was collected regularly, i.e. at least once a week, to ensure that the minimuim number of degenerating spermatozoa were present in the semen sample, thereby providing samples as uniform as possible for analysis [El Jack and Lake, 1966 b]. Semen was collected from twenty-nine males in and twenty-seven males in on many separate occasions and pooled each time; about 20 min was required to accomplish each pooled collection. The semen from each cock was transferred, immediately after collection, into a centrifuge tube kept in a bath of water at 10 C. The pooled semen was mixed thoroughly but gently with a Pasteur pipette by drawing semen in and out, and finally by rolling the tube between the hands. Precautions were taken to avoid frothing during the transference and the mixing of the very viscous semen from the vasa deferentia otherwise air bubbles became trapped in the semen. Aliquots of whole semen for analysis were taken at this stage before the spermatozoa could settle out. Determination of the amount of trapped plasma in packed spermatozoa and the true volume of spermatozoa in semen. - To determine the content of ions per unit volume of spermatozoa it was first necessary to estimate the amount of trapped plasma in packed spermatozoa after centrifugation, since the content of ions in whole semen and seminal plasma fractions is readily analysed, and from these parameters the content of ions inside the spermatozoa was calculated. The following formulae were used to estimate the trapped plasma and hence the true volume of spermatozoa in semen samples: Vs (/B)l (1) V =Va -Vs (2) V ==Total volume (ml.) of spermatozoa. Va =Apparent volume (ml.) of packed spermatozoa after centrifugation Vs =Volume (ml.) of trapped seminal plasma in packed spermatozoa.

3 334 El Jack and Lake Vd =Volume (ml.) of diluting fluid added to cells after centrifugation to resuspend the trapped plasma. A=g N2 per ml. seminal plasma. B =g N2 per ml. diluted, trapped seminal plasma. The following procedures were carried out to find out the unknowns of the formulae (Pyrex glassware was used throughout). About ml. whole semen was transferred accurately to a small, straight-sided tube. The semen was then centrifuged for 30 min (1500 g, 40 C.); the standard conditions of centrifugation were chosen after preliminary experimentation had shown that all the spermatozoa were thereby separated from the seminal plasma and no appreciable damage was sustained by the spermatozoa when they were subsequently resuspended. If the semen was spun at more than 1500 g, spermatozoa were likely to be appreciably damaged; deformed spermatozoa, as well as an increased number of stained heads, were observed in nigrosin-eosin smears of the resuspended semen using the method of El Jack and Lake [1966 b]. The speed of the centrifuge was tested frequently throughout the 2-year-period during which analyses were performed. The levels of the upper meniscus (total semen volume) and the apparent volume of spermatozoa (Va, Formula 2) were marked carefully on the tube after centrifugation for subsequent calibration with a Grade A pipette. The apparent volume of spermatozoa was determined by this method because the amount of semen available for each of the separate sets of analysesin this study was limited and insufficient to fill simultaneously a Wintrobe haematocrit tube. Great care was taken after marking the tube not to disturb the spermatozoa whilst the seminal plasma was taken off with a Pasteur pipette and transferred to another tube for use in subsequent analyses of seminal plasma. It was next necessary to resuspend the trapped seminal plasma in a diluting fluid and so the inner wall of the tube containing the spermatozoa was wiped carefully with a filter paper and the small amount of seminal plasma left behind on top of the packed cells was absorbed. The wall of the tube was then washed by introducing cold (50 C.), isotonic sodium citrate (3.55 per cent) solution drop by drop down the side of the tube till it was almost full; it was quickly extracted to leave the spermatozoa undisturbed. The inside of the tube was again dried with a filter paper making sure that all the citrate was removed from just abo-ve the surface of the mass of spermatozoa. 2-5 ml. (Vd, formula 1) of cold sodium citrate (3.55 per cent) solution was added to the tube and the packed spermatozoa resuspended thoroughly using a wide-mouthed Pasteur pipette, making sure that no large aggregates of spermatozoa remained. It was then centrifuged for 30 min (1500 g, 40 C.) and the supernatant removed. To complete formula 1 the nitrogen contents in aliquots of the neat seminal plasma and that diluted with the known volume (2.5 ml.) of sodium citrate were determined. Duplicate 0-1 ml. aliquots of neat plasma and 0 4 ml. of diluted plasma were taken for the estimation of nitrogen by the method of Chibnall, Rees and Williams [1943] after digestion in microkjeldahl flasks using copper-selenate as catalyst. Determination of the content of potassium, sodium, magnesium and calcium in seminal plasma and whole semen. - Duplicate aliquots (0 5 ml.) of whole semen were accurately measured into 10 ml. stoppered tubes. 4-5 ml. 01 N HCI was added and after mixing thoroughly was left to stand overnight at room temperature. It was then centrifuged and the supernatant transferred to a clean tube. This extract was used for sodium, potassium, calcium and magnesium estimations after aliquots had been diluted 10-fold (calcium), 50-fold (magnesium), 400-fold (potassium) and fold (sodium) with doubly glass - distilled water. Aliquots (0 5 ml.) of seminal plasma were treated likewise except that only 50-fold dilutions were made for the estimation of potassium. The concentrations of calcium, magnesium, potassium and sodium were determined

4 Composition of Fowl Semen 335 by atomic absorption spectrophotometry [El Jack and Lake, 1966 a]. In view of the observations of Willis [1960, 1961] that the presence of appreciable amounts of phosphorus in some biological material tended to interfere with the estimation of calcium, a preliminary test was carried out to determine the extent of any phosphorus interference. None was detected and thus samples directly diluted with distilled water were considered suitable for analysis. Determination of the content of chloride in seminal plasma and whole semen. - 0 I ml. aliquots of both seminal plasma and whole semen were accurately measured into plastic pots and the chloride content estimated by an electrometric titration method [El Jack and Lake, 1966 a]. Determination of the content of carbon dioxide in seminal plasma. - Semen was collected under liquid paraffin and transferred into small centrifuge tubes containing liquid paraffin which were immersed in a cold water bath to minimise the loss of CO2 to the atmosphere. Within a few minutes, the semen was centrifuged (1500 g, 4 C.) for 15 min and the plasma extracted and kept under liquid paraffin. The plasma was analysed within 30 min of collection from the males. As semen was collected by the massage technique, and so was momentarily in contact with air, a test was made to find out whether there was any detectable loss of CO2 to the atmosphere. For this purpose cocks were anaesthetized and semen was withdrawn from the vas deferens exposed by an incision lateral to the cloaca. A 2 ml. syringe sealed with liquid paraffin was used. No difference was found in the CO2 content of seminal plasma collected by the two methods and thus routine collections of semen by massage were considered satisfactory for use. Carbon dioxide was estimated by gas chromatography in a similar manner to that described for blood plasma [El Jack and Lake, 1966 a]. Determination of the water content of spermatozoa and seminal plasma. - An appropriate volume of clean semen was centrifuged (1500 g, 40 C., 30 min) in a tared tube and the seminal plasma removed; a measured volume of the latter was placed in another tared vessel. The part of the tube above the packed spermatozoa was carefully cleaned and dried. The vessels containing seminal plasma and packed spermatozoa were weighed and dried at 105 C. overnight. Afterwards they were cooled in a desiccator and weighed to constant weight using a Mettler semi-microbalance. The air in the weighing chamber of the balance was kept as dry as possible by placing a number of small containers of silica gel about the pan of the balance. Estimation of the density of spermatozoa in semen samples. - Semen was diluted 60-90,000-fold with red cell diluting fluid (Fisher Chemicals, Ltd.) and the spermatozoa counted with an electronic Fisher Autocytometer. Ashing procedure to extract inorganic ions. - Preliminary experiments were carried out to ascertain whether ashing was necessary before ions could be extracted from the various fractions of semen. Aliquots of whole semen and seminal plasma were ashed in platinum crucibles (450 C., overnight) and afterwards cooled and treated with five drops 6N Nitric acid (B.D.H. Analytical grade). The samples were heated to 200 C. to get rid of the nitric acid and a few drops of hydrogen peroxide were added to complete the ashing. It was found that the simpler, direct, acid-dilution method gave the same results as obtained after the ashing procedure and thus the former method was adopted routinely for the determination of cations. This finding was confirmed by Quinn et al. [1965]. Ashing was not a suitable technique for chloride determinations because the element disappeared during the procedure. RESULTS The average volume of uncontaminated semen obtained from a Brown Leghorn cock was 0-15 ml. and the density of spermatozoa was 5-72 x 106 per cu.mm. The volume of packed spermatozoa under the present centri-

5 336 El Jack and Lake fugation conditions (1500 g, 30 min, 4 C.) was 20-7 per cent of the total semen volume; the amount of plasma trapped within the spermatozoan pellet was 72-2 per cent of the packed cell volume (S.E. of the mean + 146; twenty-one observations), so that 5-6 per cent of the volume of semen was accounted for by the spermatozoa. A check was carried out on the magnitude of the estimate of trapped seminal plasma, obtained under the conditions of centrifugation in the present study, when Dr. S. Iverson, Beatty Memorial Cancer Hospital, Glasgow, calculated the volume of a fowl spermatozoon as 9-2,um3 with a model of the Coulter counter which also estimated size. From his information and the above-mentioned figures for the density of spermatozoa in semen samples, it can be calculated that the trapped plasma could be expected to occupy 73 per cent of the packed spermatozoan volume which was in excellent agreement with the mean figure of 72-2 per cent estimated routinely by the nitrogen method. The water content of uncontaminated seminal plasma was 96-4 per cent (S.E.M ) and that of spermatozoa was 59-9 per cent (S.E.M ). The content of water in seminal plasma was higher than that found in comparable cock blood plasma (95 per cent; Gilbert, private communication) which may be partly accounted for by the lower protein content of seminal plasma (0.8 per cent) compared with blood plasma (5.0 per cent). The amount of water in spermatozoa was lower than that in erythrocytes (70 0 per cent; Gilbert, private communication). The specific gravity of spermatozoa ( ) exceeded that of erythrocytes (1.09; Gilbert, private communication) and in this respect resembled the spermatozoon of the mammal [Mann, 1964]. There is extremely little cytoplasm in the spermatozoon and the high specific gravity is most likely due to the highly condensed nuclear TABLE I. THE AMOUNTS (M-EQUIV./L., MEAR±S.E.M.) OF THE PRINCIPAL INORGANIC IONS IN UNCONTAMINATED SEMINAL PLASMA, WHOLE SEMEN AND SPERMA- TOZOA OF THE BROWN LEGHORN COCK AND THE CONTENT OF CARBON DIOXIDE IN SEMINAL PLASMA. Electrolyte Seminal Plasma Whole Semen *Spermatozoa Sodium ±205t (21) ±1-67 (21) Potassium ±0-28 (21) 15-60±0-261 (22) Calcium ±0 06 (17) 2-46±0-12 (18) 0-72 Magnesium.. 511±0-15 (21) 5-80±0101 (23) Chloride ±1-31 (16) 41-64±1-46 (16) Carbon dioxide ±1-12 (10) * The amount of ions in spermatozoa was calculated from the amounts in seminal plasma and whole semen after determining the volume of trapped plasma (72 2 par cent) in packed spermatozoa. Numbers in parentheses refer to the number of pooled samples of seminal plasma and whole semen analysed. t =P <0-01 and t =P <0-001 in a comparison between whole semen and seminal plasma. material, the aggregation of mitochondria in the midpiece and the fibrous nature of the tail. The results in Table I showed that seminal plasma, in common with most extracellular fluids, contained more sodium ( m-equiv/l.) and calcium

6 Composition of Fowl Semen 337 (2x55 m-equiv/l.) than the spermatozoa (53x58 and 0-72 m-equiv/l. respectively). The spermatozoa, like somatic cells, contained more potassium (61x38 m-equiv/l.) and magnesium (17x09 m-equiv/l.) than extracellular fluid, i.e. seminal plasma (12.93 and 5-11 m-equiv/l. respectively). The concentration of chloride in the seminal plasma (41-89 m-equiv/l.) appeared to be very similar to that inside the spermatozoon (37.22 m-equiv/l.). The carbon dioxide content of seminal plasma was m-equiv/l., similar to that of blood plasma [El Jack and Lake, 1966 a]. The contents of ions in the spermatozoa were calculated from the amounts in whole semen and seminal plasma, the volume of trapped plasma in packed spermatozoa, and the proportion of cells to plasma in semen. The data for sodium, calcium and chloride in Table I indicate that the amounts of these ions inside the spermatozoon are small in comparison with the seminal plasma. It was felt that this conclusion was further supported by the fact that in a few of the individual calculations of the content of these ions in spermatozoa, derived from the amounts found in the corresponding whole semen and seminal plasma fractions, a small negative amount of these ions was obtained. DIscussIoN The content of the principal inorganic ions in individual samples of seminal plasma from the vasa deferentia was very consistent considering that the samples of semen were collected and studied at intervals over a period of 2 years. This is probably not surprising since, unlike mammals, there are no accessory reproductive organs which contribute large and variable amounts of fluids to semen. Also, the cocks were massaged frequently for semen collection throughout the entire period of study with rests between collections of no longer than 7 days. This practice tends to produce semen of uniformly good quality [El Jack and Lake, 1966 b]. In spite of possessing little cytoplasm, the spermatozoon of the fowl contained a relatively large amount of potassium in comparison with that in the seminal plasma. This agrees with what is known about the general composition of the nucleus and cytoplasm in somatic cells [Siebert and Humphrey, 1965]. The amount of potassium present in the spermatozoon was less than that found in the erythrocyte [El Jack and Lake, 1966 a] which may be due to the greater amount of cytoplasm in the latter cell. These findings for potassium in spermatozoa agree with those of Quinn et al. [1965]. The actual amount reported by the latter was slightly less but this may be accounted for partly by their having analysed the packed cells which would include trapped plasma. The relatively large amount of magnesium present in the spermatozoa is most likely associated with the mitochondria aggregated in the midpiece. It is felt that although the amounts of sodium, calcium and chloride were found to be smaller in the spermatozoon than in the seminal plasma further studies are required to substantiate the absolute amounts. Difficulties are

7 338 El Jack and Lake encountered in techniques of estimating intra-spermatozoan contents, especially of sodium chloride and calcium, one reason being the relatively large amounts of these ions in the surrounding plasma. No significant difference was demonstrated in the levels of calcium or chloride between seminal plasma and whole semen, and the content of sodium in seminal plasma was significantly higher than in whole semen (P < 0-01). It is inadvisable to use washed cells since washing is likely to lead to a leakage or entry of unknown quantities of ions. Unlike us, Quinn et al. [1965] reported a very large amount of sodium and calcium in fowl spermatozoa and it is considered possible that this discrepancy is once again due to their having analysed packed cells containing trapped plasma, which itself contains much sodium. On the other hand, Siebert and Humphrey [1965] found much sodium in the nuclei of liver cells (160,u-equiv./g dry wt). It is notable that the amounts of potassium and magnesium in seminal plasma are much higher, and calcium lower, than those in blood plasma [El Jack and Lake, 1966 a]. In the case of magnesium this difference may be associated with the apocrine type of secretion occurring in the male reproductive tract whereby large portions of cells are extruded into the lumen. Measurements of the amounts of ions within spermatozoa in general are sparse, and what is available tends to show great variability between species. How much of this is due to varying techniques, true variation between species or variation within a species caused by other factors are interesting questions for future consideration. Keitel and Jones [1956] reported that the seminal plasma and spermatozoa in human semen were of similar composition; there was a high concentration of sodium and chloride and a low concentration of potassium in both compartments. However, they reported the presence of prostatic calculi of unknown composition in the packed cells used for analysis. Cragle, Salisbury and van Demark [1958] found a larger amount of potassium in bull spermatozoa than in seminal plasma, but Crabo [1964] found a smaller amount of potassium in the spermatozoa. Steinbach and Dunham [1961] investigated the amounts of sodium, potassium and chloride in the spermatozoa of some invertebrates (Arbacia, Mytilus and Phascolosoma) when trapped extra-cellular fluid in the pellets was determined with inulin. It was shown that spermatozoa resembled other cells in maintaining high internal potassium and low sodium and chloride concentrations relative to those in sea water. Quinn and White [1966] demonstrated variations in semen cations in the ram, bull and human in relation to semen quality and methods of collection. The small amount of chloride in seminal plasma from the vas deferens of the fowl agreed essentially with previous findings [Lake et al., 1958; Lake and El Jack, 1964 a], but is in complete disagreement with the high amount reported by Hammond et al. [1965]. The latter obtained very large volumes of semen from males (2-07 ml.) and it is possible that there was contamination with transparent fluid from cloacal glands. The chloride concentration they found in their samples of seminal plasma was about the same as that in blood

8 Composition of Fowl Semen 339 plasma [Lake and El Jack, 1966 a, b) which is known to be the origin of the chloride in transparent fluid. The small amount of chloride found in the seminal plasma of the vas deferens is interesting from the point of view of the ionic balance in this medium in which spermatozoa mature and survive for at least 2 weeks. It has been found that glutamate assumes the role of principal anion in place of chloride [Lake and El Jack, 1964 b). The fact that fowl spermatozoa mature in an environment in which the level of chloride ions is low may be one of the causes for the deterioration of spermatozoa in diluents containing approximately blood levels of chloride [Lake, 1960; El Zayat and van Tienhoven, 1961]. REFERENCES BLYTH, J. S. S. (1954). 'Notes on the Poultry Research Centre flock of BrownLeghorns', Wld's Poult. Sci. J. 10, CHIBNALL, A. C., REES, M. W. and WILLIAMS, E. F. (1943). 'The total nitrogen content of egg albumin and other proteins', Biochem. J. 37, CRABO, G. L. (1964). 'Concentration of sodium, potassium and chloride in the epididymal plasma and sperm of bulls', Proc. 5th Int. Congr. Anim. Reprod. and A.I. Trento. Sec CRAGLE, R. G., SALISBURY, G. W. and VAN DEMARK, N. L. (1958). 'Sodium, potassium, calcium and chloride distribution in bovine semen', J. Dairy Sci. 41, EL JACK, M. H. and LAKE, P. E. (1966 a). 'The distribution of the principal inorganic ions in venous blood of the adult domestic cock and the content of carbon dioxide in the plasma', Brit. Poult. Sci. 7, EL JACK, M. H. and LAKE, P. E. (1966 b). 'The effect of resting roosters from ejaculation on the quality of spermatozoa in semen', J. Reprod. Fert. 11, EL ZAYAT, S. and VAN TIENHOVEN, A. (1961). 'Effect of chloride ions on cock spermatozoa', Amer. J. Physiol. 200, HAMMOND, M., BooNE, M. A. and BARNETT, B. D. (1965). 'Study of glucose, electrolytes, enzymes and nitrogen components of fowl seminal plasma', J. Reprod. Fert. 10, KEITEL, H. G. and JONES, H. S. (1956). 'The mineral and water composition of normal human sperm', J. Lab. clin. Med. 47, LAKE, P. E. (1957). 'Fowl semen as collected by the massage method', J. agric. Sci. 49, LAKE, P. E. (1960). 'Studies on the dilution and storage of fowl semen', J. Reprod. Fert. 1, LAKE, P. E. (1966). 'Physiology and biochemistry of poultry semen'. In Advances in Reproductive Physiology. Vol. 1., Ed. A. McLaren. London: Logos Press Ltd. LAKE, P. E., BUTLER, E. J., MCCALLUM, J. W. and MACINTYRE, I. J. (1958). 'A chemical analysis of the seminal and blood plasmas of the cock', Q. Jl exp. Physiol. 43, LAKE, P. E. and EL JACK, M. H. (1964 a). 'Further observations on the chemical composition of the seminal plasma of the domestic cock', Proc. 5th Int. Congr. Anim. Reprod. and A.I., Trento. Sec. 2, LAKE, P. E. and EL JACK, M. H. (1964 b). 'The origin and composition of fowl semen'. In Physiology of the Domestic fowl, pp Eds. C. Horton-Smith and E. C. Amoroso. Edinburgh: Oliver and Boyd Ltd. MANN, T. (1964). In The Biochemistry of Semen and the Male Reproductive Tract. London: Methuen. MuNRo, S. S. (1938). 'The effect of testis hormone on the preservation of sperm life in the vas deferens of the fowl', J. exp. Biol. 15,

9 340 El Jack and Lake QUINN, P. J. and WHITE, I. G. (1966). 'Variation in semen cations in relation to semen quality and methods of collection', Fert. Steril. 17, QuiNN, P. J., WHITE, I. G. and WIRICE, B. R. (1965). 'Studies of the distribution of the major cations in semen and male accessory secretions', J. Reprod. Fert. 10, SIEBERT, G. and HUMPHREY, G. B. (1965). 'Enzymology of the nucleus', Adv. Enzymol. 27, STEINBACH, H. B. and DUNHAM, P. B. (1961). 'Ionic gradients in some invertebrate spermatozoa', Biol. Bull. 120, TAKEDA, A. (1959). 'Studies on the cock semen. I: Na, K and Ca levels of the seminal plasma', Res. Rep. Fac. Text. Sericult. Shinohu Univ. 9, WmLIS, J. B. (1960). 'The determination of metals in blood serum by atomic absorption spectroscopy. I: Calcium', Spectrochim. Acta 16, WILLIs, J. B. (1961). 'The determination of calcium and magnesium in urine by atomic absorption spectroscopy', Anal. Chem. 33,