Background: To our knowledge, there are no publications that have evaluated physical activities in relation to the etiology of AIS other than sport scoliosis (1,2) so we undertook a study to assess the physical activities of patients with progressive AIS from their first year to early teens and compared these with those of a control group.

Methods: All 156 children in this study had to adhere to the following 6 criteria:- born full term, fed well as infants, achieved their milestones, no hospital visits except for sports injuries, no family history of a scoliosis and no back pain (prior to diagnosis in the patient group). We compared 79 consecutive patients (girls 66, boys 13) diagnosed as progressive AIS (62 of whom subsequently had a spinal fusion) with a control group of 77 subjects (girls 66, boys 11) of similar age, gender, race and socioeconomic status. A structured history was obtained from the mother and child of each group average time taken to obtain the history was 47 minutes. Each child was examined for toe touching and vertical symmetry of spinous processes whilst standing. The findings suggest a relation between physical activities or the lack of them and the development of progressive AIS.

Results: There is a significantly increased odds of AIS in those who were introduced to a swimming pool within the first year of life (p=0.001), did not attend gymnastics/ karate classes (p=0.005), did not attend dance classes (girls only, p=0.045), did not have horse riding classes (p=0.003), did not go skating (< 0.001), and who could touch their toes (p=0.011). No association is found with playing football/hockey or regular swimming at the age of 10 years.

Conclusion: Progressive AIS is positively associated with an early introduction to swimming and ability to toe touch. Spinal asymmetry was noted in the controls. AIS is negatively associated with participation in dance, skating, gymnastics/karate and horse riding classes. Is it possible that children who develop AIS have a longstanding proprioception defect which makes them less likely to participate in sporting activities? If so, by encouraging children to participate in sport might we increase their proprioception abilities and make those at risk less likely to develop spinal asymmetry which may progress to a scoliosis requiring surgical correction?

Objective. Despite the current revolution in molecular medicine that has benefitted the treatment of certain diseases (Ross 2002), idiopathic scoliosis has resisted attempts to understand the molecular basis of its curve development. Lowe et al (2002) in a longitudinal study of 55 AIS patients concluded that platelet calmodulin levels correlate closely with curve progression and stabilization by bracing or spine fusion. They suggest that the platelet is a “minimuscle” with a protein contractile system (actin and myosin) similar to that of skeletal muscle. Using Lowe’s data we found that percentage platelet calmodulin change correlates significantly with percentage Cobb angle change (ANOVA, p=0.0003, n=54) that led us to suggest a platelet/skeletal hypothesis to account for their findings as part of a cascade concept for the pathogenesis of AIS.

Hypothesis. The human immature vertebral body is unusual among mammals in lacking epiphyses. This may explain why an axial load transmitted directly from the intervertebral disc deforms mature vertebral body end-plates as an axial inward bulge (Brinckmann et al 1983). In immature normal vertebral bodies vascular “lakes” (resembling bunches of grapes) have been found adjacent to the disc growth plates in subjects aged 9 to 13 years of age (Mineiro 1965). These “lakes” may provide a susceptibility to platelet activation from vascular stasis and shear stresses. In addition to their role in hemostasis platelets contain many growth factors including TGF-βs in α-granules that are secreted at a fracture site (Bolander 1992). TGF-βs are found in human neonatal rib growth plates (Horner et al 1998) but, like estrogen receptors, have not yet been sought in human intervertebral disc growth plates. We hypothesize that in the presence of a small scoliosis curve (from unknown causes ?spine, rib, muscle, or nervous system) platelets, as they circulate through vessels in eccentrically-loaded and deforming immature vertebral bodies particularly about the curve apex in the presence of a basic defect, are activated also by repeated axial inward bulges of disc growth plates causing mechanical micro-insults with endothelial cell desquamation and the formation of a calcium-cadmodulin complex. The latter is associated with platelet contraction (shape change) and the secretion from α-granules of various growth factors including angiogenic regulators (platelet release reaction, Hartwig 2003, Reed 2002, Rendu and Brohard-Bohn 2002). These growth factors abet the hormone-driven growth of the already mechanically-compromised disc growth plates and induce anterior spinal overgrowth and curve progression. The basic defect in AIS could be 1) a platelet, endothelial, or subendothelial anomaly, defect, or functional (?hormonal) disorder, and 2) one or more genetic polymorphisms that involve platelet receptors (Afshar-Kharghan and Bray 2002) and putative estrogen receptors in vertebral disc growth plates (Inoue et al 2002ab). The predilection of progressive AIS for girls may be related to the cyclical platelet functions in women associated with normal uterine function (Jones et al 1983, Pansini et al 1986, Tarantino et al 1994, Faraday et 1997). Curve laterality is determined by factors that initiate curve progression. Low plasma melatonin of progressive AIS may act both by a reduced antagonism to calmodulin (Lowe 2000, Dubousset and Machida 2001) and facilitating platelet aggregation with secretion of growth factors from α-granules.

Conclusions. The platelet/skeletal hypothesis for progressive AIS and the cascade concept suggests much new research. The hypothesis has genetic, diagnostic, prognostic and potential therapeutic implications. It raises questions about the possibility of changes in platelet calmodulin levels in other progressive and resolving deformities that occur in the immature and adult skeleton.

Objective. To evaluate the relation of ribs to the spine in the transverse plane (TP) at the curve apex in preoperative AIS using a real-time ultrasound method and radiographs (Burwell et al 2002).

Design. With the subject in a prone position and head supported, readings of laminal rotation (LR) and rib rotation (RR) were made on the back by one of two observers (RKA, ASK) using an Aloka SSD 500 portable u/s machine with a veterinary long (172mm) 3.5 MHz linear array transducer. The maximal difference between LR and RR about the curve apex was calculated as the apical spine-minus-rib rotation difference (SRRD). The SRRD eliminates the effect of any anterior chest wall asymmetry on the ultrasound measurements and, assuming no movement of ribs in the TP at the costotransverse joints, is considered to be a measure of TP rib deformity. The radiographic Cobb angle (CA), apical Perdriolle rotation (AR), and apical vertebral translation (AVT) were measured by one observer (RGB). In an attempt to separate mechanical axial vertebral rotation from axial vertebral deformity a derivative was calculated as Perdriolle rotation minus ultrasound LR with the latter corrected for the positional effect of lying prone and termed the axial vertebral difference (AVD) The correction factor (CF) used is maximal Scoliometer angle of trunk rotation obtained in the standing forward bending position minus that in the prone position.

Subjects. Thirty-three preoperative patients with AIS were studied (thoracic curves 20, thoracolumbar curves 8, double curves 5).

Results. The mean figures in degrees or mm (AVT) are shown in the Table.

All curves combined. The LR is significantly greater than the RR (p< 0.001) and correlates with RR (r=0.358, p=0.041), SRRD (r=0.713, P< 0.001) but not with CA (p=0.088), AR (p=0.166), AVT or AVD. AR does not correlate significantly with CA.

Thoracolumbar and thoracic curves. In the thoracolumbar curves the SRRDs are significantly greater than those in the thoracic curves (p=0.031) implying more TP rib deformity in the thoracolumbar curves. In the thoracic curves the SRRDs correlate negatively with the AVDs (r= −0.470, p=0.036) suggesting that rib deformity and intravertebral deformity contribute reciprocally and together with axial spinal rotation to determine the overall spinal deformity of AIS.

Conclusions. The findings are consistent with the hypothesis that in preoperative AIS the axial RR and TP rib deformities are adaptations to rotational and lateral forces imposed by the scoliotic spine (Wever et al 1999). Might surgical stiffening of the posterior ends of the apical convex ribs – in an attempt to prevent TP convex rib deformity – constrain axial spinal rotation, vertebral translation and intravertebral deformity and limit curve progression? #Supported by AO/ASIF Research Commission Project 96-W21

Nachemson (1996), drawing upon the theses of Sahlstrand (1977) and Lidström (1988), articulated the view there are more girls than boys with progressive AIS for the following reason. The maturation of postural mechanisms in the nervous system is complete about the same time in boys and girls. Girls enter their skeletal adolescent growth spurt with immature postural mechanisms – so that if they have a predisposition to develop a scoliosis curve, the spine deforms. In contrast boys enter their adolescent growth spurt with mature postural mechanisms so that they are protected from developing a scoliosis curve. There is evidence that postural sway improves with age in boys and girls until about 10 years of age after which it is similar between the sexes (Hirashawa 1973, Odenrick and Sandstedt 1984) findings which need further evaluation. We term Nachemson’s concept the neuro-ossesous timing of maturation (NOTOM) hypothesis. It may have an evolutionary basis through natural selection towards sexual and skeletal development during adolescence being earlier in girls and later in boys.

The NOTOM hypothesis suggests a treatment to prevent progression of late-juvenile idiopathic scoliosis, early-AIS, and some secondary scolioses based on delaying the onset of puberty used therapeutically in girls with idiopathic precocious puberty (IPP, Grumbach and Styne 1998). The proposal is to administer a gonadorelin analogue which in the pituitary down-regulates the receptors to hypothalamic gonadotropin-releasing hormone (GnRH) causing a fall in both luteinizing hormone (LH) and follicle-stimulating hormone (FSH), which in turn causes a fall in oestrogens and androgens, and thereby delays or stops menarche and slows bone growth – as in girls and boys with IPP (Galluzzi et al 1998). Expert scrutiny of this therapeutic proposal is currently in progress.

1. This paper reports a histological study of the fate of sheep and calf cancellous bone grafts impregnated with autologous red marrow of Wistar rats and implanted intramuscularly as composite xenograft-autografts for two to twelve weeks. It also includes some biochemical estimations of certain types of sheep and calf bone used to prepare these composite grafts.

2. Only one of 223 devitalised bone xenografts implanted without autologous marrow formed new bone; in contrast 216 of 223 transplanted with marrow formed new bone.

3. The new bone formed by the composite grafts is derived from the autologous marrow. There was no evidence for an inductive effect upon the marrow of the various types of xenograft bone studied as described previously for allograft bone (Burwell 1966).

4. The highest score of new bone formation was found in composite grafts based on fully deproteinised sheep iliac bone prepared at Oswestry. Statistically this score was significantly higher than those registered by composite grafts prepared from intact (frozen and freeze-dried), decalcified (frozen and freeze-dried) and Kiel sheep bone, and by Kiel and Oswestry calf bone (Table II).

5. The histological evidence reported suggests that the high score with the sheep Oswestry composite grafts is because Oswestry bone is feebly immunogenic, if at all; and that such feeble or absent immunogenicity permits more marrow cells to differentiate into osteoblasts and lay down new bone without impediment.

6. The lower scores of new bone formation in most of the undeproteinised composite grafts of sheep origin–intact frozen, intact freeze-dried and Kiel–are attributed to residual immunogenicity within the organic material of the donor bone, because each type evoked the formation of mature plasma cells.

7. The Kiel bone grafts appeared to evoke less of a plasma cell reaction and may be less immunogenic than the intact and decalcified bone xenografts.

8. The sheep Oswestry CXA's formed significantly more new bone than did the calf Oswestry CXA's. This difference may be due to the different physical properties of the bone obtained from old sheep compared with the bone obtained from a young calf.

1. The present study is an attempt to analyse and apportion significance to the role of inductive mechanisms in bone transplantation.

2. The experimental model used in the present work is that of the composite homograftautograft of cancellous bone previously described (Burwell 1964a).

3. Iliac bone was removed from hooded rats and washed free from its marrow. The bone was then treated by various physical and chemical methods (some of which have been used by other workers to prepare bank bone), namely freezing (-20 degrees Centigrade, -79 degrees Centigrade, -196 degrees Centigrade); freeze-drying (without sterilisation, sterilisation with high energy radiation, sterilisation with ß-propiolactone); decalcification (with E.D.T.A.); irradiation (in the frozen state at a dose of 4 million rads); boiling in water; immersion in merthiolate solution; extraction of organic components with ethylenediamine: and calcining at 660 degrees Centigrade. The treated bone was then impregnated with fresh autologous marrow procured from the femoral shaft of the Wistar rat into which the treated composite graft was to be implanted. The grafts were inserted intramuscularly and removed for study after two, six and twelve weeks.

4. After fixation, serial sectioning and staining, each graft was examined microscopically, and the proportion of new bone/grafted bone scored using an arbitrary scale (0-4). The mean score (and the standard error of the mean score) was then plotted for each treated composite graft and also for several types of fresh cancellous bone grafts.

5. It was found (Fig. 2) that the various treated composite grafts formed a spectrum of bone-forming capacities–the maximum scores being attained by the frozen and freeze-dried composite grafts, the lowest scores by the "deproteinised" composite grafts.

6. The reasons for these differences are discussed. It is concluded that cancellous bone, after transplantation, has the property to induce and promote osteogenesis in marrow; moreover, that this property is contained in the organic components of bone.

7. From the standpoint of inductive mechanisms, cancellous bone treated by freezing or freeze-drying seems to be the most suitable devitalised bone for grafting purposes; bone which has been boiled or merthiolated less suitable; and "deproteinised" bone the least suitable.

8. Freeze-dried bone sterilised physically (by high energy radiation) or chemically (by ß-propiolactone) did not form significantly less new bone than did freeze-dried bone which had not been sterilised.

9. Remodelling mechanisms in bone transplantation are briefly discussed and attention drawn to the deficiencies of present knowledge. The quantitative studies of other workers have indicated that freeze-dried bone may be more rapidly remodelled than is frozen bone.

10. The importance of fresh red marrow in promoting osteogenesis in bone transplantation and in the healing of certain fractures, is emphasised. It seems likely that the interrelationship of bone and marrow revealed by experiment has wider significance not only in health and in response to injury but also in causation of certain idiopathic bone disorders.

1. Previous immunological studies have shown that homografts of fresh marrow-free iliac bone are only weakly, if at all, antigenic.

2. In view of this finding an attempt was made to produce a foreign bone graft capable of forming new bone as readily as an iliac autograft by the following method. Living cells of high osteogenic potential and of autologous type were introduced into the graft by combining homologous fresh marrow-free iliac bone with the animal's own red marrow to form a fresh composite homograft-autograft of cancellous bone.

3. Such fresh composite homograft-autografts were inserted into a muscular site in Wistar rats and removed for microscopical examination at intervals of one to seven days and at two, six and twelve weeks after transplantation.

4. It is found that bone and marrow together as a fresh composite homograft-autograft form considerably more new bone than do either of the components of the graft transplanted separately. Homografts of fresh marrow-free iliac bone form, in general, a small amount of early phase and late phase new bone. Autografts of red marrow transplanted alone to a muscular site formed new bone in thirteen to thirty experiments (43 per cent).

5. The stimulus to osteogenesis, and the cellular source of osteoblasts, in marrow autografts is discussed in the light of present knowledge. The concept is suggested that after its transplantation there develops in marrow an inductive system leading to osteoblastic differentiation and bone formation. It is proposed that the necrosis of a portion of a marrow graft liberates osteogenic substances which are taken up by primitive wandering cells derived from littoral cells lining the vascular sinusoids of the surviving portions of the marrow which are induced, thereby, to differentiate as osteoblasts.

6. The cellular source of osteoblasts in a fresh composite homograft-autograft of cancellous bone is discussed. It is deduced that the new bone is derived mainly from the contained marrow of the graft, by mechanisms similar to those leading to osteoblastic differentiation in transplanted autografts of marrow.

7. The stimulus to the greater formation of new bone by fresh composite autograft-homografts than by autografts of marrow transplanted alone is discussed. Two explanations are suggested: 1) a more extensive necrosis of marrow in a composite homograft-autograft than in marrow transplanted alone; and 2) an inductive effect of bone upon marrow.

8. The new bone formed by autografts of fresh marrow-containing iliac bone, it is concluded, is derived not only from osteoblasts on the surfaces of the grafted bone but also from primitive wandering cells derived from littoral cells lining the vascular sinusoids of the surviving portions of its marrow.

9. Mechanisms which may play a role in the histogenesis of woven bone are discussed.

10. The significance of the relation of bone and marrow is considered briefly in the light of knowledge concerning the venous patterns of bone and marrow.

1. The antigenicity of homologous cortical and cancellous bone has been investigated in eighty-four rabbits.

2. The primary immune responses which occur in lymph nodes draining homografts of fresh tissues (Burwell and Gowland 1961, 1962) have been used as a histological indicator of the antigenicity of fresh homologous cortical bone freed from soft tissues.

3. The secondary immune responses which occur in lymph nodes draining homografts of fresh marrow-containing iliac bone (Burwell 1962a, b) have been used also as a histological indicator of the antigenicity of homografts of 1) fresh cortical bone freed from soft tissues, 2) fresh marrow-free iliac bone, and 3) mairow-containing iliac bone treated by boiling, freezi ng, freeze-drying and merthiolate solution.

4. It is found that whereas fresh homologous cortical bone fails usually to produce cytological evidence of a primary response in the regional lymph nodes, fresh homologous cortical bone chips inserted into the drainage areas of lymph nodes sensitised previously to donor ..tissue evoke constantly cytological evidence of a secondary response.

5. Fresh homologous marrow-free iliac bone inserted into the drainage areas of lymph nodes sensitised previously to donor tissue does not produce detectable evidence of a secondary response.

6. Homografts of boiled marrow-containing iliac bone do not elicit a secondary response in lymph nodes previously sensitised to donor tissue.

7. Previous work has shown that homografts of frozen (–20 degrees Centigrade) marrow-containing iliac bone do not evoke a primary response in lymph nodes draining such grafts. In the present work it is shown that similar frozen homografts inserted into the drainage areas of lymph nodes previously sensitised to donor tissue evoked a secondary response in three of six lymph nodes.

8. Homografts offreeze-dried marrow-containing iliac bone fail usually to evoke a secondary response in lymph nodes sensitised to donor tissue.

9. Homografts of marrow-containing iliac bone treated by immersion in merthiolate solution before being inserted into the drainage areas of lymph nodes previously sensitised to tissue from the donor elicited a secondary response in three of five lymph nodes.

10. Knowledge concerning the antigenicity offresh and treated homologous bone is discussed in the light of recent work.

1. Experiments to examine the antigenicity of homologous bone tissues in rats are reported. The tissues studied included fresh marrow-free cortical bone blocks and chips, fresh, boiled, frozen and freeze-dried marrow-containing iliac bone, fresh iliac bone devoid of marrow, and fresh red marrow.

2. The various tissues were transplanted from hooded to Wistar rats. Three weeks later a skin graft from each donor was transplanted to its respective host to detect the presence of transplantation immunity, which was indicated by the early rejection of the skin graft.

3. Homografts of fresh cortical bone evoked transplantation immunity indicating that it contained transplantation antigens which were also in the skin.

4. Homografts of fresh marrow-containing iliac bone also evoked transplantation immunity, which was shown to be caused by the red marrow.

5. Fresh iliac homografts devoid of marrow did not elicit transplantation immunity. This suggests that iliac bone tissue may not contain transplantation antigens or that the small amount of iliac bone inserted was insufficient.

6. Microscopy of the grafts, removed after three weeks, showed that the inflammatory infiltrations around the bone homografts and autografts were not very different, but that the amount of new bone formed was different. The autografts produced a lot of new bone, the homografts only a little.

7. It is suggested that the immune response evoked in the host by the foreign graft impairs the formation of new bone by fresh homografts of cortical blocks, cortical chips and marrow-containing iliac bone.

8. The impairment of new bone formation by homografts of marrow-free iliac bone is discussed. Such bone grafts fail to evoke detectable transplantation immunity. Why these grafts do not form more new homologous bone than the other homografts studied, is not clear.

9. Homografts of boiled and frozen iliac bone do not evoke any detectable change in the sensitivity of the host to donor tissue.

10. Homografts of freeze-dried marrow-containing iliac bone elicit a slight but significant prolongation of the survival of skin homografts. The implication, in terms of modern immunological theory, is that in such grafts certain tissue antigens still persist.

1. The response of the first regional lymph node to a homograft of fresh iliac cancellous bone inserted subcutaneously into the rabbit's ear three weeks after the introduction of a similar graft from the same donor into the same ear has been investigated in thirty rabbits. Fifteen rabbits which received second-set autografts of cancellous bone have also been studied.

2. The insertion of second-set homografts of fresh marrow-containing cancellous bone evokes an immune secondary response in the lymph nodes draining the grafts.

3. The increase in weight of the first regional lymph nodes on the side receiving second-set homografts is more rapid and of greater magnitude than that of nodes draining first-set homografts of cancellous bone. Second-set autografts evoke weight changes in the draining nodes similar to those in nodes draining first-set autografts of cancellous bone.

4. The histological changes which occur in the lymph nodes draining the second-set homografts (secondary response) are described and compared with those occurring in lymph nodes draining first-set homografts of cancellous bone (primary response).

5. In the primary response the distribution of large and medium lymphoid cells is throughout an activated sector of the cortex of the lymph node (Burwell and Gowland 1961), but in the secondary response these cells are found peripherally within the activated sector of the node. In both the primary and the secondary responses large and medium lymphoid cells are found in the medullary trabeculae of the lymph nodes.

6. The differences between the primary response of lymph nodes draining a tissue homograft (cancellous bone) and the primary response of lymph nodes draining classical antigens, and reported by other workers, are described.

7. Knowledge concerning the inflammatory response in the tissues of the host surrounding homografts of fresh cortical and cancellous bone implanted into animals previously sensitised to tissue from the respective donor is reviewed.

8. The late phase of new bone formation by homografts of fresh cancellous bone is discussed in the light of immunological studies.

1. The antigenicity of cancellous bone has been investigated in ninety-seven rabbits.

2. The immune responses of lymph nodes draining fresh homografts of cancellous bone (Burwell and Gowland 1961b) has been used as a histological indicator of the antigenicity of components of fresh homologous cancellous bone and also of the antigenicity of homologous bone subjected to a variety of physical or chemical treatments.

3. The principal antigenic component of a fresh homograft of iliac cancellous bone is the nucleated cells of the red marrow.

4. Homologous marrow-free cancellous bone does not usually produce cytological evidence of an immune response in the lymph node draining the graft, unless new homograft bone formation occurs.

5. The treatment of marrow-containing cancellous bone by boiling, freezing at - 20 degrees Centigrade, freeze-drying, irradiation or by merthiolate solution impairs the transplantation antigenicity of the tissue as a homograft.

6. The immersion of cancellous bone in a glycerol-serum-Ringer solution which is then slowly cooled to - 79 degrees Centigrade, stored for one week and then rapidly thawed, allows considerable preservation of the antigenicity of the red marrow.

7. Knowledge concerning the antigenicity of fresh and treated homologous bone is discussed.

8. Evidence is presented to show that the large and medium lymphoid cell response of lymph nodes draining homografts is due principally to the T-antigens, rather than H-antigens, of the grafts.

9. The changes which occur in the first regional lymph nodes draining tissue homografts may provide another test system to assess the transplantation antigenicity of foreign tissues or extracts of foreign tissues other than bone.

1. The effects of the insertion of pieces of fresh cancellous bone into the subcutaneous tissues of the ear upon lymph nodes and spleens have been investigated in seventy rabbits.

2. The main immunological response is found to occur in the first regional nodes draining the sites of insertion of homografts of bone, which show a considerable increase in weight compared with nodes draining autografts of bone.

3. An increased number of large and medium lymphoid cells occurs principally in the first regional node of the homografted animals, as Scothorne and McGregor (1955) observed using skin as the homografted tissue.

4. The large and medium lymphoid cell response is found in both the cortex and the medulla of the lymph nodes. In the cortex a sectoral distribution of the cellular response is observed and the name reactive cortex is given to these sectors. Evidence is presented to show that the sectoral pattern of reactivity is probably determined by the localised entry into the node of iso-antigens through lymphatic vessels draining the bed of the graft.

5. We have made a quantitative analysis of the large and medium lymphoid cell response in the reactive parts of the diffuse lymphoid tissue of the cortex. The mean maximal large and medium lymphoid cell response occurs five days after the insertion of bone homografts.

6. The origin and fate of the large and medium lymphoid cells and their role in the production of antibodies is reviewed in the light of recent work.

7. A correlation is made between the maximal production of large and medium lymphoid cells in the first regional lymph node, the invasion of the graft bed with small lymphocytes and the inhibition of new bone formation in the homografts.

1. An immunological examination of the sera of thirty rabbits which had received primary and secondary homografts of cancellous bone into a subcutaneous site did not reveal the presence of circulating precipitins, haemagglutinins or passive haemagglutinins. These findings are consistent with the observations of Bonfiglio and his colleagues (1955).

2. Electrophoretic examination of the serum of four rabbits receiving primary and secondary homografts of bone into an intramuscular site did not reveal any change in the serum protein fractions.

3. A search for auto-antibodies produced by primary and secondary autografts of cancellous bone was unsuccessful in fifteen rabbits.

4. The multiple injections of saline extracts of bone into four rabbits did not evoke the production of demonstrable circulating antibodies, results which are in accord with the findings of Bonfiglio and colleagues (1955) and Curtiss and colleagues (1959).

5. For the first time the production of classical antibodies in response to injections of extracts of heterologous bone has been recorded. The repeated injections of a saline extract of rabbit bone intraperitoneally into ten mice produced demonstrable precipitins and passive haemagglutinins both to protein and polysaccharide fractions present in the bone extracts.

6. Knowledge concerning the production of humoral antibodies to transplants and extracts of bone has been reviewed.