180 there is no hypercapnic stimulus for bronchodilatation and asthmatics have no protection. In addition, it has been noted that asthmatic subjects usually hyperventilate out of proportion to their CO2 production during exercise, whereas healthy subjects do not. Although some workers believed that hypocapnia could be relevant9 they could not demonstrate this-perhaps merely because an increase in CO2 acts a2 a more powerful bronchodilator than a decrease in CO2 does as a bronchoconstrictor. Because end-tidal COZ tensions have not, as far as I am aware, been measured in asthmatics while swimming (although they have been well documented in healthy subjects in sports mediae journals) the potentially important protective property of hypercapnia may have been overlooked. Institute of Respiratory Medicine, Royal Prince Alfred Hospital, Camperdown, NSW 2050, Australia PETER M. DONNELLY 1. Reggiani E, Marugo L, Delpino A, Piastra G, Chiodini G, Odaglia G. A comparison of vanous exercise challenge tests on airway reactivity in atopical swimmers. J Sports Med Phys Fitness 1988; 28: 394-401 2 Bar-Yishay E, Gur I, Inbar O, Neuman I, Dlin R, Godfrey S. Differences between swimming and running as stimuli for exercise-induced asthma. Eur J Appl Physiol 1982; 48: 387-97. 3. Kohrt WM, Morgan DW, Bates B, Skinner JS Physiological responses of tri-athletes to maximal swimming, cycling and running. Med Sci Sports Exercise 1987, 19: 51-55. 4. Cordain L, Stager J Pulmonary structure and function in swimmers. Sports Med 1988; 6: 271-78. 5. Town GP, Vanness JM. Metabolic responses to controlled frequency breathing in competitive swimmers Med Sci Sports Exercise 1990; 22: 112-16. 6. Anderson SD. Is there a unifying hypothesis for exercise-induced asthma? J Allergy Clin Immunol 1984; 73: 660-65. 7. McFadden ER Jr. Exercise-induced asthma as a vascular phenomenon. Lancet 1990; 335: 880-83. 8 Asmussen E, Neilsen M. Studies in the regulation of respiration in heavy work Acta Physiol Scand 1946; 12: 171-78. 9. Silverman M, Anderson SD, Walker SR. Metabolic changes preceding exercise- induced bronchoconstriction. Br Med J 1972; i 207-09. What causes motoneuron disease? SiR,—Your Oct 27 editorial is correct to emphasise the epidemiological aspects of motoneuron disease (MND). Although the ultimate truth may be found by molecular biologists, the epidemiological approach, in the short-term, is more likely to give clues as to cause and prevention. The ubiquinated inclusions found in neurons of patients with MND and other neurodegenerative disease provide a marker for a sick neuron. However, Dr Swash and Dr Martin (Dec 1, p 1379) fail to mention that such inclusions are present in cells outside the nervous system and may be no more important than lipofuscin. The importance of genetic factors in MND can be assessed by twin studies. We have identified 120 pairs, of whom only 2 are concordant. In a recent study of 90 pairs,’ none was concordant. There are only 2 other well-documented twin pairs and in both instances they were dizygotic and concordant.z3 The evidence, therefore, points strongly to an environmental cause4,5 such as a virus4 or chemical agent. 5 Supported by the Motomeurone Disease Association. Department of Neurology, Ipswich Hospital, Ipswich IP4 5PD, UK C. H. HAWKES A. J. GRAHAM 1. Currier RD, Conwill DE. Influenza and physical activity as possible nsk factors for amyotrophic lateral sclerosis: a study of twins. In: Rose FC, Norris FM, eds. Amyotrophic lateral sclerosis new advances in toxicology and epidemiology. London: Smith Gordon, 1990 23-28. 2. Dumon J, Macken J, de Barsy TH Concordance for amyotrophic lateral sclerosis m a pair of dizygous twins of consanguineous parents. J Med Genet 1971, 8: 113-16. 3. Estnn WJ Amyotrophic lateral sclerosis in dizygous twins. Neurology 1977; 27: 692-94. 4. Kennedy PGE On the possible role of viruses m the aetiology of motor neurone disease: a review. J R Soc Med 1990; 83: 784-87. 5. Hawkes CH, Cavanagh, JB, Fox AJ. Motoneuron disease. a disorder secondary to solvent exposure? Lancet 1989; i: 73-75. SIR,-Dr Swash and Dr Martin rightly emphasise the importance of molecular pathology and genetics in the investigation of the cause of motoneuron disease (MND). Although 90% of cases are sporadic with no family history of the disorder, familial cases constitute the remaining 10% and can be adult onset, autosomal dominant, and associated with dementia in some families. However, juvenile-onset cases are also seen with duration commonly over 10 years and spasticity, from pyramidal tract involvement, as the main clinical feature. Inheritance may be dominant or recessive and there has been no report of both juvenile and adult onset types within the same family. 1 The gene locus for childhood proximal spinal muscular atrophy (where degeneration is restricted to the anterior horn cells of the spinal cord) has been localised to 5q.2 As Swash and Martin point out, this finding provides a step towards understanding genetically determined degeneration of anterior horn cells. It also provides a mo’°1 for investigating familial cases of MND, especially those of juvenile onset There is an urgent need therefore to obtain and store DNA samples from arose uncommon familial cases for such studies. European Neuromuscular Centre, 3743JN Baarn, Netherlands ALAN E. H. EMERY 1. Emery AEH, Holloway S Familial motor neurone diseases. Adv Neurol 1982; 36: 139-45. 2. Melki J, Sheth P, Abdelhak S, et al Mapping of acute (type 1) spinal muscular atrophy to chromosome 5q12-q14. Lancet 1990; 336: 271-73. SIR,-We strongly support your editorial about the unequal geographical distribution of motoneuron disease (MND). Our epidemiological studies substantiate this finding. We have shown that the worldwide increase in MND mortality is also taking place in both England and Wales and the Republic of Ireland.’ In addition, we found that MND mortality in the South African white population was a half of that expected at England and Wales rates and that men from the Asian subcontinent who immigrated to England had half the mortality, and females a fifth of the mortality from MND compared with that of England and Wales. In contrast to Dr Swash and Dr Martin we agree with Dr Critchley and Dr Mitchell’s statement (Dec 1, p 1380) that "any ethnic differences might yield new insight". Supported by the Motoneurone Disease Association. Charing Cross Hospital. London W6 8RF, UK Medico-Social Research Board, Dublin, Republic of Ireland MARTA ELIAN GEOFFREY DEAN 1. Elian M, Dean G In Rose FC, Norris FM, eds Amyotrophic lateral sclerosis new advances in toxicology and epidemiology. London: Smith-Gordon, 1990. Long-term survival in "corrected transposition" SiR,—The natural history of corrected transposition of the great arteries (C-TGA) without significant associated cardiac anomalies remains poorly defined because few adult cases have been reported. 1-5 We describe two men aged over 60 with uncomplicated C-TGA, in whom the diagnosis of the transposition was made on the basis of non-invasive diagnostic techniques, including calor doppler echocardiography, magnetic resonance imaging (MRI), and intravenous digital subtraction angiography (DSA). A 67-year-old man without symptoms was referred to hospital for evaluation of electrocardiographic (ECG) abnormality. He had a single, accentuated second sound at the upper left sternal border. A faint holosystolic murmur (grade 1 /6) was heard clearest at the lower left sternal border. ECG showed first-degree atrioventricular block with no septal Q wave in leads I, aVL, V5, and V6, and inverted T waves in V3-V6, I, and aVL. The cardiac outline was moderately enlarged, and the ascending aortic shadow was not prominent in the right upper mediastinal border. Calor doppler echocardiography, MRI (figure), and DSA showed typical findings for C-TGA; the aorta arose from the left-sided (morphological right) ventricle with heavy trabeculations, and the medially placed pulmonary artery arose from the right-sided (morphological left) ventricle with no intracardiac abnormalities. Echocardiography revealed diffuse slight hypocontractility of the left-sided ventricle with mild mitral (anatomical tricuspid) regurgitation. A 60-year-old man, also referred because of an ECG abnormality, had no symptoms. Cardiac examination revealed a soft holosystolic murmur (grade 2/6) at the apex. ECG showed left-axis