This article was written in 1994 for the academic journals but was not accepted for publication
CONSERVATISM IN THE DEEP NORTH: Trends in Queensland attitudes
University of N.S.W.
and G.P. Hall
I.Q. Solutions Australia
Previous surveys carried out in 1980 confirmed the stereotype of Queenslanders as being more conservative than other Australians. It was hypothesized that this could be an outcome of Queensland's more Northerly and hence warmer climate. The 1980 surveys were closely replicated in 1994 but with the addition of scales to measure environmentalism and racism. After controls for the influence of demographic differences were carried out, however, there were now found to be no significant differences between Queenslanders and New South Wales residents. This change is possibly related to the downfall of the Conservative governments that once dominated Queensland.
There has been considerable comment from many quarters over the years to the effect that Queensland is a "different" State from other Australian States and that this difference consists primarily of Queenslanders being more conservative or even "red-neck" (Ray, 1982a & b, 1983a & b).
This reputation seems in part to derive from the many years during which Queensland was led by Premier Sir Johannes Bjelke-Petersen. There can be no doubt that "Joh" was for many years one of the most Right-wing figures on the Australian political scene. "Joh" is however now long gone from State politics and Queensland has for some years now been led by the very pragmatic Labor Party Premier Peter Beattie. Mr. Beattie is nominally a socialist but his government would probably be most informatively described as "middle-of-the-road". In the United States he would probably be described as a "fiscal conservative" -- meaning that he has some welfarist policies but nonetheless keeps total spending to a figure within the current State income. Whatever he is, however, he is clearly far less Right-wing than "Joh".
It is interesting, therefore, to ask whether the continuing popularity and hold on power of Mr Beattie signifies some sort of sea- change in the attitudes of the average Queenslander. Against that view, it might be argued that not much has changed in Queensland -- as many of the policies adopted by Mr Beatty could be seen as a continuation of "Joh" policies -- e.g. his fiscal rectitude and his opposition to prostitution and marijuana use. So the political scene does have ambiguous implications as far as Queensland public opinion is concerned.
Queensland is in the Northern part of Australia and the wider interest of the Queensland scene lies in the analogy with the American "Deep South". The Southern States of the United States are, like Queensland, climatically warmer than the rest of the country and known for Red-neck and conservative attitudes. If there is some effect of climate on attitudes, therefore, it would be reasonable to expect in Queensland attitudes similar to those of the American "Deep South".
The Queensland climate is warmer because, being in the Southern hemisphere, it is, like the American South, nearer to the equator. That an analogy is commonly perceived between the two regions is embodied in the occasional characterization of Queensland as Australia's "Deep North" and in the fact that Queenslanders themselves often refer to immigrants from the more Southerly Australian States as "Mexicans".
A public opinion survey in 1980 showed that the stereotype of Queensland conservatism did have a basis in fact at that time (Ray, 1982a & b,1983a & b) so it is of interest to know how enduring this has been. Was there a deep-seated (climatic?) difference between Queenslanders and others that has survived the test of time or was the difference found in 1980 merely due to various temporary circumstances (such as the Red-baiting and scaremongering of "Joh")? A replication of the 1980 survey in 1994 seemed therefore worthwhile.
A replication would not only allow an examination of contemporary differences between State populations but would also allow an examination of any changes within the States over time. As any sort of longitudinal or diachronic study is rare, the opportunity to conduct one on the present occasion seemed well worth pursuing.
The 1980 survey which showed an attitudinal difference between Queensland respondents and respondents from the more Southerly State of New South Wales was a postal (mail-out) survey. It achieved a 31% response-rate from a one-wave mailout and was shown to give a demographic distribution virtually indistinguishable from contemporary doorknock surveys. An attempt was made, therefore, to replicate this methodology as closely as possible. Sampling was therefore done by random selections from the Australian Electoral Rolls (i.e. Federal voter registration lists).
One difference, however, was that the original study was done under the aegis of the University of N.S.W. whereas the present study was done under the aegis of a market-research firm (IQ Solutions Australia). This difference was of course reflected in the letterhead used for the questionnaire and there was some concern that a commercial letterhead could reduce the response-rate. It was hoped, however, that a fully-printed and well-designed covering letter might convey a reasonable impression of the importance of the project.
Queensland is a more decentralised State than N.S.W. so it seemed that the more rural character of Queensland's population would have to be taken into account as an influence on attitudes. One way to allow for this possibility is to compare capital cities only. Both Brisbane and Sydney are substantial metropolises with populations of roughly 1 million and 3 million people respectively so neither is in any objective sense rural. The capital city sub-samples of the overall sample were therefore of considerable importance in their own right.
To allow for the fact that Brisbane accounts for a smaller proportion of the State population, however, 700 survey forms were sent out in Queensland compared with 500 in New South Wales. It was hoped that this would give capital-city sub-samples of roughly comparable size. Exactly the same strategy was adopted in the 1980 survey.
The attitude scales used to measure conservatism were also exactly as in 1980. There were 3 scales -- to measure General Social Conservatism, Economic Conservatism and Moral Conservatism. Economic Conservatism was separated out in deference to Lipset's (1959) contention that conservatism on economic issues is far more politically crucial than conservatism on social issues generally. The economic conservatism scale, then, focused on issues of government intervention in the economic system and welfarism versus private enterprise and individual effort. Such issues might seem more national than regional but they were nonetheless once regularly focused on in Queensland by "Joh". Issues of sexual morality were also separated out in the form of the Moral Conservatism scale as such issues have long been prominent in Queensland political life. The scale focused, then, on such issues as sex-education, homosexuality and pre-marital sex. The items of all three scales have been listed in Ray (1983a).
One of the great political issues that was not included in the 1980 survey was the environment. To remedy this, the present survey included the 12 item version of the Australian Environmentalism scale (Ray, 1974 & 1975). This scale had been empirically selected from a large pool of items and had shown a reliability (Cronbach's alpha) of .78. The scale had been found to correlate .317 with a scale of then- topical political conservatism issues so seemed possibly relevant to the present concerns.
It seemed possible, however, that the scale might have dated somewhat in the 20 years since its construction so to allow for this four items on undoubtedly current environmental topics were also included in the questionnaire.
Another important lacuna of the 1980 study was its failure to include measures of racial attitudes. Prejudicial attitudes towards blacks and other races are of course an important part of the usual characterization of the American Deep South (the Ku Klux Klan etc.) so an examination of any comparable attitudes in Queensland seemed important. To this end the questionnaire included scales of Attitude to Asians and Attitude to Aborigines (Australian native blacks). These two groups were judged to be the most likely prejudice objects in contemporary Queensland. The scales were versions of those used earlier in Ray & Lovejoy (1986) and the most recent versions are listed here.
A response-rate of 27% was achieved in both States (ns of 196 in Qld. and 135 in NSW). This is a little down on the 31% achieved in 1980 and could be accountable for as either the result of a lot of surveying by many parties in the interim or as reflecting the fact that the 1980 result was unusually good. Other Australian postal surveys conducted by the senior author in the 1970s typically had response-rate figures in the middle 20s (e.g. Ray, 1980, Study II etc).
As part of the questionnaire, respondents were asked what part of the State they lived in -- with a precoded choice being given of the capital city or elsewhere in the State. This allows respondents to be classified in terms of where they see themselves as living.
It might be noted that this is unlikely to coincide well with official geographic designations. Sydney, for instance, is officially only one small municipality to the East of a vast urban sprawl that includes many other municipalities. A person living in the municipality of Penrith (to the West of the same urban sprawl) might (or might not), however, still describe himself as living in Sydney. Sydney, in other words, in colloquial use tends to give its name to the conurbation of which it forms the oldest part but where the borders of this perceptual conurbation are could not be exactly specified. The Brisbane conurbation is much less fragmented but the same problems arise nonetheless.
There were then 135 New South Wales respondents of whom 52 saw themselves as living in Sydney. There were 196 Queensland respondents, of whom 84 saw themselves as living in Brisbane.
There were three notable demographic biases in the sample: The proportion of males in the sample was nearer two thirds than half (but did not differ significantly between the two States); the Sydney sub- sample was significantly older (t = 3.01) than the Brisbane sub-sample and the Brisbane sub-sample was significantly more likely (t = 2.14) to vote for the Labor party than was the Sydney sample. The latter is not necessarily a bias at all in that Queensland was at the time of writing the only State government that was in Labor party hands and it is in the cities rather than in the country that the Labor party has its strength.
The sex and age biases are more troubling but can of course be controlled for statistically. As such biases did not emerge with the 1980 mail-out sample, it can only be presumed that the use of a commercial rather than a university letterhead was the crucial factor in bringing about the sex differences observed.
The age difference could also be partially attributable to the letterhead factor but is also to a small degree explainable by reference to what has happened to Brisbane demographics as a result of the massive immigration from interstate that has been taking place for many years now. Routine telephone surveys taken by IQ Solutions Australia in South East Queensland on a monthly basis always reveal that roughly half the population in that area are not Queensland born. As South East Queensland includes around half of the State population, it may be seen that Queenslanders could in the forseeable future become a minority in their own State.
The 1980 results were all scored in 3, 2, 1 format -- meaning that all degrees of agreement with any given item were scored as 3 and all degrees of disagreement were scored as 1. Other responses or omitted responses were scored as 2. This procedure reduces the discriminating power of the individual item and hence the reliability (alpha) of the scale. Nonetheless, for the sake of comparability, the same procedure was followed on the present occasion.
The reliabilities (alpha -- see Cronbach, 1951) of the scales on the combined sample of 331 Queensland and N.S.W. respondents were: General Social Conservatism .69; Economic Conservatism .50; Moral Conservatism .82; Environmentalism (with 4 new items) .57; Asians .78 and Aborigines .83.
Some of these reliabilities are quite low so the reliabilities of the scales when each item is scored in 5,4,3,2,1 format (i.e. scored to reflect degrees of agreement with each item) are of some interest. These were: General Social Conservatism .76; Economic Conservatism .54; Moral Conservatism .85; Environmentalism .65; Asians .81; Aborigines .88.
Even in these circumstances, then, the Economic Conservatism and Environmentalism scales had unsatisfactory internal consistency -- indicating that attitudes in these areas are no longer organized as they once were. The widespread collapse of the more extreme forms of socialism (i.e. the European "Communist" regimes) is the obvious explanation for the de-polarization of economic attitudes that is revealed in the Economic Conservatism scale: State enterprise is just no longer generally seen as a plausible alternative to capitalism. As Economic Conservatism was the one form of conservatism that did NOT differ between the States in the 1980 survey, however, the failure to measure it adequately on the present occasion is not of great moment.
More surprising is the loss of coherence in environmental attitudes since 1974. The original 12-item scale had a reliability of .57 under 5,4,3,2,1 scoring and the 4 new items bought this up to .65. It would seem therefore that the issues that were central around the time that the environmental movement rose to prominence are much less central today. Thus while the four new items helped to give the augmented scale some usefulness, the wisdom of hindsight would indicate that even more such items should have been included.
The Mean Scores
The mean scores of the Queensland and New South Wales samples are given for all scales in Table 1. It will be seen that all differences were small but that there was nonetheless a significant (p <.05) tendency for Queenslanders to be more anti-Aboriginal (anti- black).
As was found in the 1980 study, the capital city results were generally very close to the Statewide results but the significant differences between the capitals were that the Brisbane people were more permissive on issues of sexual morality and were more pro- environment than were the Sydney people.
Means (and S.D.s) for two States and their capital cities. "Gen" = General Social Conservatism; "Eco" = Economic Conservatism; "Mor" = Moral Conservatism; "Env" = Environmentalism; "Abo" = Attitude to Aborigines; "Asi" = Attitude to Asians; "Edu" = Educational level; "Res" = Resident. Differences that are statistically significant <.05 are starred.
.........N.S.W. Resident Qld Resident Sydney Resident Brisbane Resident
Gen.... 66.32 (8.04)....... 66.28 (7.22)... 66.05 (8.28)...... 64.58 (8.19)
Eco.... 22.74 (3.85)........ 22.68 (3.70)... 22.92 (4.34) ..... 22.88 (3.46)
Mor.... 26.98 (6.55)........ 26.69 (6.69)... 28.05 (6.02)...... 25.79 (6.69)*
Env.... 36.23 (4.76) ........36.57 (4.78) ...35.21 (5.65) ......36.95 (4.38)*
Abo.... 26.20 (5.93)........ 27.74 (6.38)* ..26.30 (6.24)..... 26.66 (6.49)
Asi..... 11.91 (3.38)....... 12.15 (3.75).... 11.94 (3.64)...... 11.56 (3.69)
Edu...... 2.70 (1.05) ,,,,,,,,,2.61 (1.11)...... 2.73 (1.09)........ 2.91 (1.10)
N .....135 ....................196................... 52 .....................84
Means and S.D.s of Queensland Population Segments. Differences that are statistically significant <.05 are starred.
..........Born in Q........ Not born in Q ..... Non-Tertiary ...... Educ
Gen.... 66.79 (7.50).... 65.87 (6.58)......... 67.67 (6.52)....... 62.98 (7.72)*
Eco..... 22.45 (3.90).... 23.14 (3.35) ....... 22.35 (3.85) ....... 23.48 (3.19)
Mor..... 26.49 (6.72).... 27.40 (6.69) ........26.95 (6.24) ........26.06 (7.61)
Env..... 35.98 (4.77) ....37.98 (4.38)* .......36.13 (4.96) ........37.62 (4.13)*
Abo..... 28.09 (6.39).... 27.01 (6.24)........ 28.94 (6.10) ........24.86 (6.09)*
Asi..... 12.27 (3.74)..... 12.00 (3.74) ........12.92 (3.64)....... 10.34 (3.36)*
Edu....... 2.57 (1.04) ......2.82 (1.12)........... n.a..................... n.a.
N ...... 121 ................. 69 ..................... 138 .................... 58
Differences in Education?
As well as the attitude scale results discussed so far, the tables include mean scores on education (scored on a four-point scale where 1 = Primary only, 2 = Some secondary, 3 = Full secondary and 4 = Tertiary). This was done with an eye to the fact that the 1980 results showed the Queensland sample as less educated, thus giving rise to the possibility that the differences then observed might have been at least in part a product of different levels of education. In both the 1980 and the current survey, education did show a slight correlation with General Social Conservatism (-.26 on the present occasion. p <.05).
Why did the mean level of education not differ on the present occasion? Perhaps because of the strong modernizing tendencies that Queensland has undergone in the last 30 or more years. Queensland was once very much a backward State under almost continuous Labor Party governments from 1915 to 1957. Under these governments large sectors of the economy came into State hands or were dominated by State-run monopolies. With the accession to power of a pro-business government (which also then successfully entrenched itself) in 1957, however, Queensland began to catch up economically with the rest of Australia. Such catching up is however necessarily gradual and improvements in anything like mean educational standards take a long time to work through the system. This process was apparently incomplete in 1980 but seems reasonably complete in the present survey -- at least in terms of the four broad educational categories used.
Another influence for change that might be suggested would be the massive immigration into Queensland from Southern States that was mentioned earlier. Although the immigration effect is large, however, it appears not to be relevant to mean State education levels as the mean education level of the Queensland-born subsample on the present occasion was negligibly different from mean for the total Queensland resident sample (Compare Tables 1 and 2). The mean score on education of the 69 respondents not born in Queensland (See Table 2) generated only a slightly more substantial difference. It may be noted from Table 2, however, that those respondents not born in Qld were more pro- environment.
The 1980 difference in mean levels of education in Queensland seemed to express the lower incidence of tertiary education in Queensland at that time (Ray, 1982a). It may be noted, however, that excluding all tertiary-educated respondents from the Queensland sample on the present occasion affected mean scores on the scales very little. (See Table 2, column 3. There are no significant differences at p <.05 between Table 1, column 2 and Table 2 column 3).
The scores of the Tertiary-educated sub-sample did differ significantly on many variables (See Table 2) from the scores of respondents with sub-Tertiary education but there were proportionately too few tertiary-educated respondents to have much effect on the overall means. It seems on balance unlikely, then, that education was the crucial factor in the different scores revealed in the 1980 study.
Controlling for Sex, Age and Rural/Urban Differences
As was mentioned previously, there were sex and age biases observed in the samples. These obviously have to be controlled for before any other results can be accepted as non-artifactual.
The sex bias was the most easily disposed of. The proportion of males in the two State samples did not differ and, within the two samples, sex did not correlate significantly with any other variable. It is therefore in a sense "naturally" controlled for and needs no statistical control.
Age, however, did differ significantly between the Brisbane and Sydney samples (Sydney mean 48.1 years; Brisbane mean 39.8 years) and was correlated significantly with a variety of other variables. It is therefore a potential source of confounding. The method chosen for removing the effect on the results of the age bias was partial correlation. It was chosen because it gives a clear picture of the magnitude of any effect described. To use this method, State of residence was made a dummy variable with 1 arbitrarily assigned to Queensland and 2 to NSW. As we are at this point explicitly concerned with controlling for extraneous variables, we here focus particularly on the Brisbane versus Sydney results as this contrast controls for any effect due to the more rural nature of the overall Queensland population.
When the data was reanalyzed with non-metropolitan results discarded, State to which the questionnaire was mailed (scored 1 or 2) was found to correlate significantly (p <.05) with the following variables: Moral Conservatism (.16), Environmentalism (-.14), Vote (.17) and Age (.19).
The implications of these results, then, are that Brisbane (Qld.) respondents were more radical on issues of sexual morality, were more pro-environment, were younger and voted more for the nominally Leftist party. It will be seen that all correlations, though significant, are very low, showing that all effects are weak. It is the first three correlations, of course, that are of substantive concern. When age was partialled out, the following effects were observed: The correlation between Moral Conservatism and State of Residence reduced to .08, which was not significant <.05; The correlation between Environmentalism and State of Residence reduced to .11 -- which is also n.s. (p <.05); The correlation between Vote and State of Residence reduced to .12 -- which is also n.s. (p <.05).
Born There or Immigrant?
It might be objected that the analyses so far have missed the point. If such a large proportion of the Queensland population are interstate immigrants, might it not be much more relevant to look at what people born in Queensland (or NSW) think? To take account of this possibility, the data was subjected to a double filter: Those not resident in Brisbane or Sydney were discarded and those who were not born either in Queensland or NSW were discarded. This left an n for analysis of 98.
In this sample, where the respondent was born showed no significant correlation with any of the scales.
Given that the differences between the Queensland and NSW respondents on the present occasion were minor, it becomes particularly interesting to look at the diachronic (longitudinal) differences. What these were may be seen from Table 3.
TABLE 3: Comparison of Conservatism scale scores in 1980 and 1994.
.................... PRESENT RESULTS......... | ........ FORMER ......
QLD.. (N=196)......... NSW (N=135) .......| QLD (N=219) ...... NSW (N=158)
Gen. ..66.28 (7.22) ...66.32 (8.04)........... | 72.84 (7.56)......... 69.01 (9.49)
Eco.. .22.68 (3.70).... 22.74 (3.85) ...........| 25.19 (5.02) .........25.85 (5.49)
Mor.. 26.69 (6.69).. ...26.98 (6.55)........... | 29.53 (6.73)......... 26.41 (7.36)
All the differences between the 1980 and 1994 scores are significant <.05 for the Queensland samples but the NSW samples differ in that the Moral Conservatism scores do not differ significantly across time. The trend, then, was for a decline in conservatism across time. As the raw data for the 1980 survey was no longer available, it was not possible to examine fully the role of demographic differences in bringing about these results but it may be noted that the 1980 and 1994 Queensland samples did not differ significantly on age and that reasons have already been given above for concluding that sex and educational differences are unlikely to have had any role in explaining the observed attitude differences.
This does, however, leave urban/rural differences as a possible explanation for the differences. To examine this, Table 4 gives the scores of the capital city residents only.
TABLE 4: Capital city scores on attitude variables in 1980 and 1993.
PRESENT RESULTS............................ | FORMER RESULTS.....
.......... Brisbane............ Sydney........... | Brisbane............ Sydney
Gen..... 64.58 (8.19) ......66.05 (8.28).. | 71.92 (7.39)........ 68.78 (9.40)
Eco...... 22.88 (3.46)..... 22.92 (4.34)... | 25.24 (4.88)....... 26.26 (5.18)
Mor...... 25.79 (6.69)..... 28.05 (6.02)... | 29.41 (6.60)....... 25.70 (7.15)
N ....... 84 .................... 52 ................. |...104 ................... 98
Once again, all the differences between the Queensland and N.S.W. (i.e. Brisbane and Sydney) samples were statistically significant <.05. For N.S.W. only the economic and moral conservatism scores were significantly different <.05. Clearly, then, there were real changes over time in Queensland -- with a general move towards reduced conservatism evident. By the same token, there was no clear general trend observable in the NSW data.
With full controls for demographic variations applied, there were NO statistically significant differences between the Queensland and New South Wales respondents. There is then no effect of State as such on any of the attitude variables examined. There is then in the present data no warrant for believing that climate has any influence on attitudes.
This result conflicts sharply with what was found for the 1980 survey (Ray, 1982a & b, 1983a & b). It is obvious, then, to ask which survey was right -- the 1980 survey, the 1994 survey or both? If there has simply been a change over time, both could in fact be right.
The obvious candidates for distorting the results of both surveys are demographic differences. In both 1980 and 1994, demographic differences were observed between the Queensland and NSW samples. The effects of demographic variations were therefore quite extensively examined on the present occasion and it was indeed found that excluding the influence of two of the demographics (age and urban/rural differences) did serve to erase what looked at first like State differences.
The only demographic variable that was not adequately excluded as a potential distorting influence on the 1980 survey, however, was education and education on the present occasion was shown not to affect to any great extent the attitude means obtained. It seems reasonable, therefore, to say that education was unlikely to have been a critical factor in 1980.
This leaves us with the conclusion that demographic differences have been excluded from both surveys as an explanation of the final results reported. The implication of that conclusion is that the Queensland difference in conservatism of attitudes did once exist but has now run its course. That being so, the difference could not have sprung from deep-lying influences (such as climate) but must have sprung from temporary circumstances -- such as the self-destruction of the Queensland branch of the Labor party at the time of the great "split" of the 1950s and the subsequent effective leadership provided by two strong Conservative politicians -- Frank Nicklin and Joh Bjelke- Petersen.
It could also be argued that even the old 1915 to 1957 Queensland Labor Party branch was very conservative on social issues because of its heavily Irish Catholic character. The implosion of first Labor Party and then Country/National Party conservatism, however, left Queenslanders quite comfortable with a moderate Labor administration (firstly under Wayne Goss and then under Peter Beattie) and the values of the Queensland people are shown to be simply mainstream at the end of that process.
It will be noted that this explanation focuses on the two-way and hence perhaps not wholly democratic nature of political influence. Not only does the public influence the leadership but the leadership can also influence the public. Even in a democracy, leaders do sometimes lead.
Cronbach, L.J. (1951) Coefficient alpha and the internal structure of tests. Psychometrika 16, 297-334.
Lipset, S.M. (1959) Democracy and working class authoritarianism American Sociological Review 24, 482-502.
Ray, J.J. (1974) Environmentalism as a trait. The Planner 14, 52-62.
Ray, J.J. (1975) Measuring environmentalist attitudes. Australian & New Zealand J. Sociology 11(2), 70-71.
Ray, J.J. (1980) How many answer categories should attitude and personality scales use? South African Journal of Psychology 10, 53-54.
Ray, J.J. (1982) Climate and conservatism in Australia. Journal of Social Psychology, 117, 297-298.
Ray, J.J. (1982) Australia's Deep North and America's Deep South: Effects of climate on conservatism, authoritarianism and attitude to love. Tableaus 169, 4-7.
Ray, J.J. (1983) The workers are not authoritarian: Attitude and personality data from six countries. Sociology & Social Research, 67 (2), 166-189.
Ray, J.J. (1983) Race and climate as influences on anxiety. Personality & Individual Differences, 4, 699-701.
Ray, J.J. & Lovejoy, F.H. (1986). The generality of racial prejudice. Journal of Social Psychology, 126, 563-564.
Go to Index page for this site
Go to John Ray's "Tongue Tied" blog (Backup here)
Go to John Ray's "Dissecting Leftism" blog (Backup here)
Go to John Ray's "Australian Politics" blog (Backup here)
Go to John Ray's "Gun Watch" blog (Backup here)
Go to John Ray's "Education Watch" blog (Backup here)
Go to John Ray's "Socialized Medicine" blog (Backup here)
Go to John Ray's "Political Correctness Watch" blog (Backup here)
Go to John Ray's "Greenie Watch" blog (Backup here)
Go to John Ray's "Leftists as Elitists" blog (Not now regularly updated)
Go to John Ray's "Marx & Engels in their own words" blog (Not now regularly updated)
Go to John Ray's "A scripture blog" (Not now regularly updated)
Go to John Ray's recipe blog (Not now regularly updated -- Backup here)
Go to John Ray's Main academic menu
Go to Menu of recent writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page (Backup here)
Go to Selected pictures from John Ray's blogs (Backup here)
Go to Another picture page (Best with broadband)