These notes represent a search through research evidence in an attempt to gain new information concerning the effectiveness of intensive behavioural intervention with young autistic children.

It is acknowledged that the attempts have succeeded in re-stating and re-describing many of the issues surrounding this approach, but that further empirical data are not yet available, and one awaits still the outcome of much needed replication studies, and of the current comparative investigations under way in this country.

This further trawl through published evidence or personal experience of early and intensive behavioural treatment of autism (as set out by Ivar Lovaas and his associates) was stimulated by a request for some up-to-date information by which to determine how to deal with continuing requests upon the LEA for funding for Lovass programmes.

Much of the argument in favour of such programmes are based upon the original Lovaas article of 1987; and a series of articles published over the following few years have debated the various methodological questions and the validity of the conclusions drawn in that 1987 study.

In other words, both the arguments in favour of Lovaas programmes and those which question the value of this type of intervention have become very familiar and regularly rehearsed; and the goal was to investigate whether new evidence or opinion has emerged in more recent years by which to move the debate forward.

The following notes reflect the present writer's attempts to trace more recent information, but no significance is to be read into the order of presentation which largely reflects the order in which the material cited was obtained or read.

McCurry (1998) describes the pressure upon a total of twenty six local authorities to adopt what he describes as a controversial and expensive treatment for autistic children despite questions raised about its efficacy in a government commissioned study.

Reference is made to the pro-Lovaas lobby which argues that this technique has the best record of enabling young children with autism to lead a normal life. (However, this begs the question of how to define ‘normal’, just as the concept of ‘recovery' in the original 1987 paper was subject to questions and criticisms). Reference is also made to concerns which have been raised over the justification for paying for treatment that many regard as unproven, and to the dilemma whereby money that might be spent upon educational provision is being diverted towards legal costs.

Rita Jordan is quoted as expressing concern over the apparently ‘magical' impact of the treatment which may serve to produce a dissatisfaction towards other intervention currently offered. Nevertheless, this is countered by claims for the frequency of mainstream school placement among children who have undergone the intensive behavioural intervention. (The problem with such claims is the apparent lack of direct data concerning long-term outcomes in this country, and a reliance, instead, upon anecdotal evidence. Beliefs that Lovaas programmes have or have not led to inclusion [however defined] in mainstream schools appear to be equally lacking in clear, empirical data involving meaningftil numbers of cases).

Meanwhile, McCurry quotes from his discussion with the National Autistic Society which takes the view that the principle of early intervention, which is a significant part of the

Lovaas approach, is significant, and that the close involvement of parents is another critically important element… but the continuing problem is that the technique may be appropriate for some children but not for others, and no single approach appears universally applicable because children with autism represent a marked range in terms of the level and nature of needs and symptoms. (In other words, the outstanding goal is still that of determining by what criteria one may match a given child with a given intervention in order to achieve maximal benefit).

Rita Jordan is further quoted as highlighting the significant number of children who appear not to make improvements as a result of the intensive behavioural treatment. The issue of the hours spent on the treatment is then discussed, with the suggestion that the critical variable is the intensity rather than the specific methodology. There is the further anxiety from Jordan that the intensive behavioural approach can lead to dependence among the children concerned thus limiting rather than enhancing their access to normal day to day routines.

In any event, McCurry's article serves to highlight the continuing dilemma in that there is a shared aspiration towards helping young children with autism in the most effective way available, but there are still questions surrounding what is the most effective way for a given individual. (The fact is that these questions can only be answered through long-term observations and replicated studies but, in the meantime, a stream of cases continue to present themselves one after another, and are dealt with through legalistic processes which may address procedural issues but not the key practice issues, or simply re-iterate rather tired arguments).

The relevant section of The book by Cohen (1998) begins by noting the 'shadowy contours’ of autism, in that it may take on different shapes with different children, or within the same child at different ages. Accordingly, it is restated that there is no standard treatment for autism but Cohen refers to a history of supposed ‘breakthroughs’ that turn out not to be the great miracle originally thought, and help only a small proportion of children with autism.

In respect of behavioural intervention, There is a further acknowledgement of the likely significance of the intensity of one-to-one work with very young children; but, while it is recognised that very good results may be achieved, this approach does not bring about significant benefits for all children. The subsequent question raised is about the potential benefits and pitfalls of any strategy, especially ‘when no approach can promise the outcome that is the dream of all parents’.

Cohen argues that most forms of treatment for autism might best be described as facilitative in that they may bring about improved functioning but they do not eliminate the core features of autism. Treatments are many and various and may include auditory integration training, vitamin therapy, dietary treatment, etc., but none of these would replace educational provision.

Seeking to summarise research evidence, Cohen points out that there is no dispute about the value of eafly intervention, and she quotes the work of Harris et al (1991) who demonstrated

that one year of intensive behavioural treatment in the pre-school period can narrow the gap in IQ and language skill that separates most autistic children from their normally-functioning peers.

In respect of intensity, it is noted that Lovaas recommended 40 hours per week of intervention and found that 10 hours per week were not sufficient to achieve benefits. However, one study that partially replicated the Young Autism Project model (Anderson et al 1987) found that even an average of 20 hours a week of one to one support over a period of a year brought about significant improvement in cognitive functioning in half the sample of children treated.

Meanwhile it appears generally agreed that the young autistic child must be helped to respond to the environment, particularly to other people; but opinions differ in respect of how this is to be achieved. For example, Temple Grandin is quoted as saying that she was one of those autistic children who could be ‘jolted’ into attention by her teachers, but other children with autism would withdraw or collapse altogether in the face of sharp intrusion into their own autistic world.

Thus, some children will respond to the highly structured behavioural approach which might be regarded as very intrusive ; but other children need to be gently encouraged to interact, as is the case with the Options/SonRise approach where the adult will initiate treatment by matching the child's actions or joining in the child's activity. Evidence exists that the adult's imitation of the child's behaviour is associated with greater social responsiveness in the child, and the success of the ‘mutual imitation’ approach may be explained in terms of the very high dosage of the strategy in a setting where distractors are minimised.

Cohen cites a number of interventionist strategies, noting that success was reported in each case with the particular child in question, but asks whether the approaches would have been effective with other children including those who were older at the start of treatment, or more severely impaired, or more difficult to interest in any kind of interaction.

For example, in response to the observed need for structure on the part of young children with autism, there have developed therapeutic nurseries which provide a nurturing environment with an emphasis upon communication and engagement in play. However, once again it can only be concluded that the approach may be highly beneficial for some children with pervasive developmental difficulties if they have some understanding of speech and have begun to communicate, but it may not be relevant with more severely impaired autistic children.

In respect of the original Lovaas study, Cohen quotes the familiar criticisms including the exclusion from the project of those children deemed likely to have poor outcomes. The target group comprised children under 40 months of age if they had no speech and under 46 months if they did have some speech, and an IQ score cut-off point was involved. Nevertheless Cohen recognises the impressive achievement of the 47 per cent of the sample said to show 'recovery' ; but the problem is that there may be many parents of children who would not have been admitted to the Young Autism Project through failing to meet the criteria for age and speech and mental age, yet still regard their children as having a very good chance of achieving normal functioning.

One should also restate the questions surrounding what ‘recovery’ means, noting that it might reasonably involve more than having an IQ in the average range and being placed in a

mainstream educational setting. The validity of such questions is justified by evidence from the Douglass Centre that the applied behaviour approach can be effective in improving IQ and language development but may not bring about improvements in social behaviour or relationships.

Other concerns about the Lovaas model relate to the emphasis upon reactive behaviour at the expense of spontaneous behaviour or initiative; or to the reliance upon the various therapists who have little training beyond the short period directly concerned with the particular behavioural approach.

While there is less of an issue now than previously about the use of aversives, Cohen still believes that the Lovaas model involves a sharpness, and one notes the work of Koegel et al (1994) whose planning was influenced by the apparent lack of happiness in the children following intensive behavioural programmes and by the goal of increased interaction and communication. Their model focuses more upon motivation and involves greater choice for the child of material and rewards, while teaching in the context of play and functional activities.

Finally there is reference to those children who stay in behavioural programmes for long periods of time with the implication that it is necessary to face up to a decision about leaving programmes when it is clear that only limited progress is being achieved.

Cohen concludes by arguing that what the Lovaas ‘school’ does better than anyone else is the documentation of outcomes; and the availability of information about the progress of the children is a major factor in the preference on the part of parents for the intensive behavioural approach.

It is argued that parents are increasingly demanding and expect to see evidence for the benefit of a given approach, and if such evidence is not available (whatever the actual success of that approach), it will not be pursued.

Her final point is that one must acknowledge that interventions have tended to be only moderately effective with many children. This may reflect the complex neurobiological foundations of autism and the considerable individual differences among children all appropriately diagnosed with autism; but it may also reflect the poor implementation of programmes or a mismatch between children's needs and particular intervention styles.

A brief review article (Rimland 1998) begins with a statement of the benefits of early intervention with children with autism, particularly when the therapy is provided at home as well as at school, but asks what type of therapy is likely to be effective and how intensive it needs to be.

Reference is made to two articles which suggest that therapies do not have to be protracted or highly intensive to be effective, but that one is not likely to find any one 'right' approach to intervention.

In the first article (Ozonoff and Cathcart 1998), there is an evaluation of the effectiveness of a home programme based upon the TEACCH model where parents were advised about their children's particular cognitive and prevocational strengths and needs, and were given demonstrations of teaching techniques such that they could continue the sessions begun by the therapists.

After four months of intervention, the authors found that the target group had made an average of nearly ten months of developmental gain compared to a control group (who simply received the ‘standard’ school provision), which was particularly impressive since most children were diagnosed with mental retardation as well as with autism.

The study was cited as providing clear evidence that the implementation of a TEACCH-based home programme is beneficial in improving the cognitive and developmental skills of young children with autism. The study was also significant in demonstrating that it is perfectly possible to combine teaching methods in that many of the children attended school programmes using techniques different from the TEACCH methods adopted at home.

The other study (Sheinkopf and Siegel 1998 ) may well have been quoted in an earlier paper in this series but it is worth reiterating the finding that the behavioural intervention for young children with autism involving only 27 hours per week, continuing for only slightly more than one year, brought about significantly higher cognitive functioning than was observed in a control group. Further, the target group showed greater improvement in symptom severity.

The authors concluded that intensive home-based behavioural treatment can be implemented successfiilly without the direct support of an academic centre, but also that benefits were gained with considerably less than the forty hours per week of treatment intensity that has been commonly advocated for these behavioural programmes.

The paper completed by Lynch (1998) reviews much of the early work on treatment for children with autism, and she, too, highlights two broad strands of interventionist style.

On the one hand, there is the behavioural approach for which Ferster (1961) was among the early theorists to argue that the limited behavioural repertoires of children with autism is an inevitable result of the absence of any reinforcing value of social stimuili. On the other hand, there is the less intrusive approach whereby the therapists work towards change through acccepting the child as he or she is, and adopting a totally child-centred form of training which focuses upon encouraging interaction.

The interesting aspect of the case study described by Lynch is that it involves a seven-year-old child (Louise), uiilike the more ‘typical’ behavioural interventions following the Lovaas procedures which have focused upon pre-school children with autism. Nevertheless, Lynch is able to cite evidence from a number of studies (such as McEachin and Leaf 1995) to demonstrate the success of the methods with older children... with implications for the benefits of combining a school prograrnme with a home-based programme.

The description begins with an acknowledgement of the criticism of home-based treatment to the effect that children lack the opportunity to mix with their peers and may become socially isolated. Reference is also made to the intensity of the treatment which some might see as ethically questionable. However, Lynch notes that attempts to introduce Louise to nursery setting were unsuccessful, and that the intensive treatment reflected the desire to enable her to cope with social interaction through training in an individual setting.

The intensive behaviour therapy consisted of non-aversive methods in individual sessions of 2 or 3 hours, making up to 30 hours per week The therapy was carried out by Louise’s mother and other therapists who received training through workshops and regular meetings.

After one year, there was found to be consistent and measurable progress in the imitation of spoken sounds and words, in object identification, and in the verbal naming of objects. In addition, concentration and attention span for tasks were found to have improved from a few seconds at a time to around 5 to 10 miriutes when reassessed by a speech and language therapist.

In her discussion of the prograrrime, Lynch acknowledges that one cannot conclusively argue that the behavioural intervention was responsible for the changes observed in Louise because there were no controls within the arrangements. Nevertheless, it is noted that a child who previously could not repeat any sounds was observed to be able spontaneously to produce words, and the improvements in attention and comprehension were also impressive. It is argued that, even allowing for maturational effects, the benefits observed could legitimately be associated with the behavioural intervention.

It was possible to be flexible within the programme thus to limit potential criticism based upon repetitive or impersonal methods. For example, it proved possible to use praise as a significant reinforcer in place of food rewards or access to toys ; and it was observed that certain activities became intrinsically rewarding. Observations also highlighted the decrease in self stimulating behaviour and the improved understanding and compliance with a range of requests.

In seeking to understand progress made, it was hypothesised that the highly structured approach used in behavioural therapy is well-matched to the concrete or ‘one-track’ thinking typical of the child with autism.

Further, while recognising that some changes may relate to spontaneous development, it was observed that modification of the input was associated with accelerated progress. For example, learning increased in the area of verbal naming when actual objects were introduced instead of pictures and photographs, and Louise moved on from what had become something of a plateau.

It was concluded that the 40 hours per week of intensive therapy advocated by Lovaas is not necessarily required for improvements to occur. Good results were gained using a maximum of 30 hours of therapy a week with an average around 25.

The success of the intervention was attributed to making the tasks very simple in the beginning so that it is easy to complete them and this becomes a very rewarding for the child. Meanwhile the intensity of the approach ensures that the child is constantly occupied and distracted from self stimulation.

Lynch^ts final point is that the use of behavioural therapy has had a positive effect on the understanding and acquisition of language which in turn increases the potential for new learning. One may, thus, argue that this type of therapy can be used successfully with a child with autism who also has a severe language problem and who is well beyond the pre-school age group.

Although criticisms of the Lovaas methodology, described in his paper of 1987, have been frequently expressed, and have also been described in a previous paper in this series, it may be helpful to highlights such criticisms again especially when those individuals who are pressing for support for an intensive behavioural programme continue to base their arguments largely upon that paper.

Discussion and exchanges of information between the present writer and the principal educational psychologist of another LEA (Selfe 1998) have highlighted the following points;

• Disquiet over the use of the term recovery, especially when it is noted that the term was replaced by ‘normal functioning’ in a subsequent publication. The question is begged why these techniques have not been taken over wholesale in specialist schools if there is such a high probability of positive outcomes.

• Concern that in evaluating behavioural methods, or in planning for individual casework, impartial and replicated studies are being replaced by litigation.

• The non-random allocation of subjects to experimental and control groups.

• The difference in average ages in the Lovaas experimental group and the control group, along with the questionable validity of intelligence testing among such very young children. The validity of cognitive measures is all the more questionable when it is noted that different tests may have been used at initial and follow up assessment. For example, the Stanford-Binet scores begin at 1Q30, while the WPPSI scale begins at 1Q41. A child with profound difficulties tested on the one instrument and subsequently on the other may appear to have gained 11 IQ points.

Since these discussions, there has arisen the further concern of ‘therapy drift’, which refers to the way in which considerable differences may arise between programmes, or even within programmes over time, or to possible differences between programmes currently operating and that which was the basis for the Lovaas 1987 study. The implication is for some concern over just what it is that is being discussed, or what is being compared to what.

A general implication is the recognition that early and intensive interventions can bring about positive change, but the degree of improvement that one can predict in the case of any given child remains unresolved. One would ideally look for a range of proven early intervention programmes in order to give parents and professionals a genuine choice, as well as providing the basis for further determining, through controlled observations, possible criteria by which to match a child with an approach.

It is interesting to note that in each of the LEAs in question, where a tribunal decision has gone in favour of the of the LEA and against the funding of a Lovaas programme, the major determining factors appear to concern the availability of a viable alternative provision, along with clear evidence of the effectiveness of that provision.

In other words, it is not enough simply to criticise the empirical foundations of the intensive behavioural interventions; it is also necessary to highlight what would be offered instead and to make available evidence for the efficacy of the existing provision.

One might also stress that it may, in some cases, be entirely appropriate to support the implementation of a Lovass programme the goal is gaining clear evidence and experience by which to determine which those cases are.

Smith (1993) has reviewed a series of studies in order to highlight what are the factors which influence treatment effectiveness.

He begins by describing the consensus about the importance of early intervention, parental involvement, community focus, and the intensity of treatment.

There are many studies which have found that intervention begun in the pre-school period achieved superior outcomes than that observed when older children are involved. Reference is made to one study which showed that, even when the intervention is restricted to pre-school children, the younger subjects attain better outcomes than older subjects.

There is a note of caution in all this in that none of the studies examined adequately controlled for the possibility that the younger and older. children were disimilar in other respects from the outset. In other words, different outcomes may reflect pre-existing differences rather than unequal responses to the treatment.

Nevertheless, Smith still holds that the evidence is compelling that early intervention increases effectiveness of any treatment.

The review also highlights the evidence that children with autism achieve better outcomes when their parents participate in the treatment than when they do not. For example, the study of Lovaas et al (1973) found that children with autism tended to relapse at the end of treatment if they returned to an institutional setting rather than to their homes. No relapse was observed among those children who returned home to parents who had learned to implement behavioural techniques.

However, Smith still describes the debate concerning the most appropriate form that parental involvement should take. For example, he refers to the risk that very intensive involvement on the part of the parents may be difficult to sustain. Similarly the logic of conducting treatment in the home and in the community, rather than in some specialist clinical setting, is recognised particularly if this can provide the opportunity for the children to practisetheir new skills in a range of settings and notjust during treatment sessions.

Existing research is equivocal in that some evidence highlights difficulty for children when the treatment situation is different from the situation in which they are expected use the skills, but other studies have failed to obtain such an effect.

Although Smith's paper is now several years old, his conclusion appears still to hold. He argues that behavioural treatment has achieved its pre-eminent position by default and not by being compared with, and surpassing, alternative interventions. There appear to be many studies which highlight the effectiveness of behavioural treatment, while other interventionist approaches may be equally effective but do not have clear experimental support. It may also be the case that some interventions are ineffective or actively detrimental, and Smith refers to relationship therapies of a psychoanalytic kind as being particularly inappropriate.

Nevertheless, Smith concludes that the work of Lovass and his associates needs to be replicated (and, despite the passage of several years, this need remains unflilfilled). Meanwhile, however, on the basis of currently available evidence, he would suggest that access to behavioural treatment should remain an available option for children with autism, and work should continue to highlight and evaluate alternative approaches and to disseminate information gained.

In an editorial piece, Schopler (1998) responds to the question whether his journal will support parents who advocate for intensive behavioural therapy. His discussion recognises why parents may wish to pursue this approach, but also notes some compelling considerations why one may not wish to support the Lovaas technique as a general policy.

Schopler notes that the intensive technique still has some major unresolved research issues and he expresses concern that the courts are being used not simply to resolve legitimate procedural questions but increasingly to judge research questions which actually can only be resolved through further experimentation and observation.

Reference is also made to the fact that the pilot study described in 1987 has not yet been adequately replicated. In any event, there are still methodological concerns which call into question the conclusions of that study, alongside gradually documented evidence for positive improvements for a number of different early intervention programmes, notably TEACCH.

Questions surrounding the appropriate intensity are cited, and Schopler reiterates his anxiety about coercive legal action against education authorities used in an effort to implement techniques that should be tested in replication research first.

This latter point (about the manner in which interventions come to be organised) is implicit in the article by Nesbitt (2000) who notes that much of the specialist service provision has been stimulated by parent-led organisations.

Her work describes the outcome of a project to investigate provisions available in a given borough, and the experiences and expectations of parents.

The results indicate that there is a significant gap, according to both professionals and parents, in the infomation available about what services exist and how access is to be gained. There was similar concern about the lack of information available about autism or autistic spectrum disorders themselves.

The author also commented about the lack of co-ordinated data available on the number of children with autistic spectrum disorders known to different professional services within the borough. In some cases information depended upon the memory of particular individuals, and there was only a limited database concerning children's needs and no classifications specific to autism.... instead the children concerned were all classified under 'learning disability'. One recommendation was the establishment of an information database to be held by social services within which there could be incorporated a system ensuring greater specificity of diagnostic category.

Training was found to be a further area of concern with very few of the professionals working with autistic spectrum disorders in receipt of any specific training. Meanwhile the local health authority recommended the development of an autistic spectrum disorder co-ordinator post designed to highlight needs and to work towards a coherent response to needs identified.

Nesbitt concludes by reiterating the need to establish a framework within which professionals and parents can work together and exchange information. Meanwhile, there is an underlining of the importance of maintaining a system of constant evaluation of services provided with a major implication for joint working among health, education, social services, and the voluntary sector.

(This kind of conclusion and the findings upon which it is based may relate equally well to other boroughs and authorities and the implication is that of translating all the goodwill into practical systems such that professionals and parents will all be fully aware of what provisions exist and of the procedures by which they are to be accessed.

In the case of assessment by professionals, there needs to be a coherent and agreed assessment ‘battery’ along with evaluation of child progress which will 'feed' the efficiency of producing criteria for the initial allocation of children among the various anterventions available.

Once again, the issue appears to be that of informed choice. In the absence of a perceived range of provision, parents may indeed continue to press for an intensive behavioural approach which may or may not be appropriate for their particular child; but the parents may not be impressed by academic arguments about methodology and will base their demands upon what they see to be available and upon data produced by the pro-Lovaas school).

Relevant to the above thoughts is the experience in another county council described by Webster(2000).

He refers to the establishment of the county's own under-fives autism project which is being evaluated externally by a university team. In addition, a similar number of children are being funded on Lovaas programmes and this intervention is being evaluated similarly.

It is noted that no specific results of the evaluation are yet available but informal evidence is offered to suggest that the Lovaas group may make good initial progress even if this may be a due to learning appropriate test behaviours. However, Webster goes on to say that there does not seem to be evidence of long-term progress over and above what might be expected with other approaches, and none of the cases has seen any reduction over time in the level of support claimed necessary by the Lovaas supervisors. There is as yet no real evidence of significant inclusion at school and future educational provision is said to be a major issue in the majority of these cases.

In respect of an impending tribunal case, the county are concerned that the primary difficulty is that of general learning rather than autism with the implication that, had this not been so, a place on the project would have been offered as a genume alternative to Lovaas and this may or may not have been tested in the tribunal.

(Once more, it appears that informed choice among alternative provisions, and ongoing evaluation, are key features).

In response to questions posed by the present writer, Howlin (2000) notes that most of the evaluative work upon Lovaas programmes has been done by individuals who are themselves committed to this approach. There are few comparative studies, and Howlin quotes the common view that intensive behaviour programmes may be a good option but there is little evidence to prove that they are better than other programmes of similar intensity. Little is known about optimal staffing levels, time span, or intensity.

Meanwhile other types of programmes of less intensity or which are school based have not published much in the way of good evaluative studies.

Howlin also notes that Lovaas and his colleagues have produced data to indicate that a significant percentage of children do not benefit greatly from the intensive behavioural approach, although there is little means of determining which children will profit best and which will not.

She concludes that the Lovaas programme can be highly beneficial for some children, but for others a different approach may offer much more.

Lovaas himself is quoted as suggesting that the critical factor may be the level of progress within the first three or four months of the programme, particularly with regard to language. One possible implication is that of establishing a programme but arranging a thorough review after six months or so of treatment to determine if this intervention should continue.

Two studies are in progress to examine the effectiveness of intensive behavioural intervention for children with autism.

The first is being conducted by Howlin and Gould at St. George's Hospital, and will report in 2003. The study involves a comparison of 2 groups of pre-school children. One group is following home-based behavioural programmes; the other is attending nursery provision.

The second study is being completed by Mudford et al at Keele University, and will report shortly. The study is examining the progress and outcomes for UK children receiving in-home and parent-managed behavioural programmes against the Lovaas 1987 results.

It is interesting to hear from one of the authors that there are immediately evident indications of significant differences between the Lovaas supervised studies, such as that described in the 1987 paper, and programmes in this country.

In response to similar questions from the present writer, Stephen Sheinkopf(2000) confirms that there seems to have been little further research evidence published recently. Following the flurry of activity after the Lovaas 1987 publication, and the claims and counter claims about outcomes and methodology, etc., things have been quiet on the Lovaas front for the last year or two.

Sheinkopf cannot identify any (independent) replication studies by which the 1987 results can be supported or refuited, but he argues that such studies remain necessary. Similarly, he cannot locate any studies which have adequately examined predictors of treatment response.

However, Sheinkopf joins the growing school of thought which suggests that the issue of intensity may be the critical one, as much as that of actual treatment style. More treatment does seem to make a difference, but there is currently a 'hopeless' confounding of the twin issues of technique and intensity. He concludes that 40 hours per week will produce more effects than 10 hours per week, but that ‘there is a lot of room (and money) between 10 and 40 hours, or even between 30 and 40 hours'.

Despite the length of these current notes, it has to be acknowledged that there has not arisen much in the way of new or clearer data over the recent past. To a large extent, it is a matter of restating the various questions and issues that have yet to be resolved.

These issues include: