Ovine neurophysin was extracted from fresh-frozen whole pituitary glands using hydrochloric acid at pH 1.2-1.6. After salt precipitation neurophysin was separated from other extracted protein by gel-exclusion chromatography on Sephadex G.75. Ion exchange chromatography on CM-sephadex C.50 using acetate buffer in a pH gradient was employed to resolve the neurophysin fraction into three components. Contamination of the preparations by non-neurohypophysial hormones was investigated. Antibodies which were raised in rabbits against ovine and porcine neurophysin were observed to have reactivity with neurophysins of the same and different species as the source of the antigen. Soluble proteins in the neural lobes of a number of mammals were separated by electrophoresis on starch-gel and tested, by immunodiffusion, for immunological reactivity with an anti-porcine neurophysin-II serum. A number of proteins, sometimes more than three per animal, reacted positively with the antiserum. Immunofluorescence histochemistry was used to demonstrate neurophysin-like antigen in tissue slices. Initially tissue was frozen unfixed and sectioned in a cryostat prior to fixation in 95% (v/v) aqueous ethanol. subsequently an improved method was developed in which tissues were fixed in saline-formalin and embedded in paraffin and immunohistochemistry carried out on deparaffinised sections. Immunofluorescence on tissue fixed in saline formalin was more intense and cells were resolved more sharply than on slices of fresh-frozen tissue. Neurophysin antigen was observed in the magnocellular nuclei of adult rats, guinea-pigs and rabbits and also in accessory cell groups in the hypothalamus. Material reactive against anti-neurophysin sera was also present in the neurohypophysis. Furthermore, neurophysin-like material was demonstrated in axons of the hypothalamo-neurohypophysial fibre tracts, especially in guinea-pig tissue slices. Immunoreactive material in axons was more abundant in tissue from dehydrated guinea-pigs. Neurophysin-like antigen, as detected by immunodiffusion and immunofluorescence was depleted in the neural lobes of dehydrated rate. Staining by aldehyde fuchsin showed that dehydration also caused a reduction in neurosecretory material in rat neurohypophyses. In the guinea-pig neurohypophysis, neurosecretory material was observed to have a lobular distribution whereas immunofluorescence was lobular on frozen sections but of even intensity on paraffin sections. Dehydration caused a depletion of histochemically-detected material in the guinea-pig neural lobe. Immunofluorescence histochemistry enabled detection of neurophysin antigen in rat hypothalami at 17 days gestational age. Fetal calves of 130 days and rabbits of 29 days gestation were also seen to have immunoreactive material in the hypothalamus and neural lobe. Evidence for modification of neurophysin-like material in the perikaryon or in proximal parts of the axona of the guinea-pig hypothalamo-neurohypophysial system was obtained by observing the effects of different treatments of HNS tissue blocks and slices on histological stains. Elements of the hypothalamo-neurohypophysial system which were stained by immunofluorescence were not always stained by aldehyde fuchsin dye.

This thesis describes the Youth Suicide Prevention Project, the aims of which were to develop, implement, and evaluate community-based strategies to address youth suicide. The focus of the evaluation was formative and was designed to constructively influence all the components of the project. The overall goal of the project was to provide information likely to improve the health and wellbeing of young people and thereby contribute, in the long term, to a reduction in youth suicide. This project, which was undertaken by Ms Coggan over a period of 26 months, involved collaboration with the Mental Health Foundation, and included the participation of community groups and young people. While several community-based strategies to address youth suicide were developed and implemented as part of this project, only three were central components of the Youth Suicide Prevention Project. The first strategy involved the use of a questionnaire designed to obtain information on the range and level of risk-taking behaviours occurring among young people and their perceptions of health-harming activities. The information obtained both assisted a local community to develop strategies to enhance the health and wellbeing of its young people and provided information to policy-makers on the extent of risk-taking behaviour and its association with self-harm. The second community-based strategy was the development and trialing of a teaching resource for use in forms three and four. This resource (Mental Health Matters) was designed to provide students with an opportunity to acquire information and skills to better promote and protect their mental health. The final strategy involved the conduct of a community action programme focused exclusively on youth health. Included in this programme was the provision and dissemination of local information on youth health and the conduct of three workshops: i) developing a mentally healthy school; ii) improving the identification, referral and acute management of suicidal young people; and iii) a community activation workshop designed to allow a community to prioritise strategies to improve youth health. Four aspects of formative input to the Youth Suicide Prevention Project are reported in this thesis: the developmental phase, the Youth Health and Risk-taking Survey, the Mental Health Matters programme, and the Youth Health Community Action Programme. This has involved the review of literature; the use of archival data and records; attendance at meetings; participant observation; the identification of evaluation procedures and methods; the design of data collection techniques; qualitative and quantitative data analyses including consultation with a statistician; and report preparation and timely dissemination both to the participants and to the wider community. The main findings of this evaluation indicated: that adopting a range of 15-24 years for youth was not realistic; that the occurrence of youth suicide and attempted suicide was spread evenly within the North Health region; that young people and school personnel saw a need for health promotion strategies to be developed; that young people identified a need for information on how to help themselves and others in times of crisis; that young people at school would provide information on a range of risk-taking activities; that high levels of risk-taking were occurring among young people at school; that an association exists between certain risk-taking behaviour and self-harm; that the provision of local information on youth risk-taking behaviour could act as a stimulus for the development of community-based strategies to address these; that a curriculum based programme with a positive focus on mental health should, with modification, be implemented nationally; that there is a need for the wider community to have information to dispel the myths surrounding suicide and suicide ideation; and that it is possible to successfully conduct a community action project exclusively focused on the health needs of young people. Major implications identified for future projects were the importance of providing relevant and local data to enable a community to develop strategies to address youth health issues and the need to further examine the high levels of risk-taking behaviours and the association of risk-taking behaviour and self-harm. This project also showed that teachers and students supported the further development of a curriculum-based programme called Mental Health Matters. This project also highlighted the need for an ongoing process of interaction between the community and research-based knowledge to ensure that strategies, proposed by the community are able so be implemented. Finally, while there is a need to focus on youth at risk, this thesis has demonstrated a need to provide the wider community with information and resources to dispel all the myths that surround suicide and suicide ideation.

In the early vertebrate embryo, the vascular system is one of the first organ systems to form. Recently it has become evident that the development of mature, functional vessels requires not only signals derived from the endothelium itself, but a set of additional molecules that are not necessarily endothelium-specific. In zebrafish and Xenopus embryos two such tissues that are believed to secrete modulators of blood vessel assembly are the hypochord and primitive gut endoderm (PGE). These tissues intimately border the major axial vasculature. Radar/Growth/differentiation factor 6a (Gdf6a) represents a signalling molecule belonging to the GDF5, 6, 7 subgroup of the transforming growth factor-beta (TGF-β) superfamily. In the zebrafish, transcripts for radar are located in the hypochord, PGE and ventral tail mesenchyme (VTM), all tissues that border the developing axial blood vessels. This prompted an investigation into a potential role for this signalling bone morphogenetic protein (BMP) during the specification and assembly of the closely related vascular and haematopoietic systems in the zebrafish. Transient forced expression experiments confirmed an early ventralising activity for the Radar signal that resulted in the expansion of the haematopoietic/vascular compartment, known as the intermediate cell mass (ICM). However, a loss-of-function zebrafish model generated using morpholino technology demonstrated a critical requirement for this BMP signal in establishing the integrity of the axial blood vessels. Furthermore, this requirement was independent of the initial establishment of vascular patterning. Zebrafish embryos depleted of the Radar signal initiate a normal primitive circulation. However, soon after this commencement of normal flow, blood cells were observed to extravasate from the axial vasculature. Microangiography confirmed this leakage phenotype. Such an angiogenic/maturation role for Radar during vascular development was supported by a transgenic zebrafish line carrying an inducible copy of the radar gene. Homozygous transgenic embryos established a typical early circulation that became progressively restricted until no blood travelled throughout the entire embryonic tissue. In summary, the work presented in this thesis strongly suggests that Radar is involved in a signalling pathway required for establishing the integrity of the axial vessels during zebrafish development.

Aims: To investigate the effect of honey on wound healing by conducting [1] a systematic review of trials of honey in wound care and [2] a randomised controlled trial of honey-impregnated dressings for the treatment of venous leg ulcers (the HALT trial). Systematic review: Method - The Cochrane Controlled Trials Register, AMED and LILACS were searched for controlled trials using honey for any type of patient with an acute or chronic wound. Honey manufacturers and researchers were contacted, citations scrutinised and the internet searched. Data from included studies were combined where studies were sufficiently alike. Findings - 18 trials were included. Honey significantly decreased healing time compared to conventional dressings for partial thickness burns (WMD -4.7 days, 95%Cl -5.1 to -4.3), but delayed healing time in comparison to early excision and skin grafting for mixed partial and deep thickness burns (WMD 25 days, 95%Cl 17.4 to 32.6). No significant effect was found for minor acute wounds or for honey compared to silver sulfadiazine in partial thickness burns. There were no trials of honey for treating venous leg ulcers. Randomised controlled trial: Method - The HALT trial was a pragmatic, open label randomised trial. Participants received either a manuka honey-impregnated calcium alginate dressing (n=187) or usual care (n=181) for 12weeks. Both groups received compression bandaging. The primary outcome was the proportion of participants with healed ulcers at 12 weeks. Secondary outcomes were time to healing, change in ulcer area, incidence of infection, adverse events, health-related quality of life and cost-effectiveness. Findings -104 participants in the honey-treated (55.6%) group and 90 (49.7%) in the usual care group healed at 12 weeks (absolute increase 5.9%, 95%Cl -4.3% to 15.7%, p=0.3). Time to healing was not significantly different between the groups (mean difference -1.8 days, 95%Cl -7.7 to 4.1, p=0.5), nor was change in ulcer area (mean difference 0.9cm2, 95%Cl -1.4cm2 to 3.2cm2, p=0.4)incidence of infection (absolute decrease 5.0%, 95%Cl -3.1% to 13.1%, p=0.2), ulcer recurrence (absolute increase 5.2%, 95%Cl -0.4% to 10.7%, p=0.1), or quality of life. The average cost of community-based treatment per participant was higher in the honey-treated group (NZ$862 versus NZ$795). More adverse events were reported in the honey-treated group (RR 1.3, 95%Cl 1.1 to 1.6, p=0.01). More participants reported pain as an adverse event when treated with honey (RR 2.5, 95%Cl 1.5 to 4.2, p=0.0001). Interpretation: Systematic review - Honey may be an effective treatment for partial thickness burns in comparison to conventional dressings. Honey does not appear to benefit healing in other acute wounds and may delay healing in mixed and partial thickness burns compared to excision and grafting. The HALT trial - Honey-impregnated dressings did not have any significant positive effect on venous ulcer healing and were more expensive than usual care. Participants treated with honey experienced more pain than control participants.

Three members of the TGFβ superfamily, bmp9, bmp10 and gdf11 were isolated from zebrafish genomic and cDNA libraries to further elucidate the role of these genes during embryogenesis. Zebrafish bmp9 and bmp10 belong to the BMP9/10 subfamily. The level of sequence similarity between the zebrafish and mammalian BMP9/10 subfamily members prompted an investigation into the evolutionary relationships of the zebrafish genes. The orthology of bmp9 and bmp10 were examined by phylogenetic and syntenic analysis. Phylogenetic analysis grouped bmp9, DSLI, Bmp9 and BMP9 into a single clade and bmp10, Bmp10 and BMP10 into another. Zebrafish bmp9 and human BMP9 were mapped to linkage group 12 and chromosome 17, respectively, and share syntenic relationships with a number of other genes mapped onto these respective chromosomes. Zebrafish bmp10 and human BMP10 were mapped to linkage group 5 and chromosome 2, respectively, and share syntenic relationships with mcm6/MCM6 and pax8/PAX8. The phylogenetic and syntenic analyses support the orthology of bmp9 and bmp10 to human BMP9 and BMP10, respectively. The phylogenetic analysis also suggests that chick DSL1 is not a unique member of the TGFβ superfamily but is the chicken ortholog of BMP9/Bmp9. The expression patterns of bmp9 and bmp10 were analysed by RT-PCR, Northern analysis, and by whole mount in situ hybridisation. No specific expression pattern was detected for either bmp9 or bmp10 by whole mount in situ hybridisation, indicating the low expression levels of these genes. The lower than expected sequence similarity to the mammalian orthologs and the low level of expression suggest a lack of evolutionary pressure between subfamily members. Zebrafish Gdf11 shows a high level of amino acid identity over the length of the entire protein to mouse GDF11. Phylogenetic analysis grouped gdf11, Gdf11 and GDF11 into a single clade. Zebrafish gdf11 and human GDF11 were mapped to linkage group 11 and chromosome 12, respectively, and share syntenic relationships with atp5b/ATP5b, dspg3/DSPG3, hoxcb/HOXC, and plasticin/PRPH on LG11/Hsa12. The phylogenetic and syntenic analyses, therefore, support the orthology of gdf11 to human GDF11. The expression pattern of gdf11 was analysed by whole mount in situ hybridisation. Expression of gdf11 was detected in both the dorsal and vegetal tail bud progenitors and during segmentation stages, expression becomes restricted to the caudal-most chordamesoderm of the tail bud. During the pharyngula stage, dynamic expression of gdf11 was detected in neural structures including the ventral hindbrain, midbrain and the forebrain. Expression of gdf11 was absent in the tail bud of ntlb195 mutant embryos but present in neural structures, indicating that the expression of gdf11 is regulated by different factors in these tissues. The function of gdf11 was analysed by microinjection of synthetic gdf11 mRNA. Overexpression of gdf11, however, resulted in the severe dorsalisation of the embryo, probably due to activation of a Nodal specific pathway.

Background: ‘Bronchiectasis’ is usually a progressive disease defined as bronchial dilatation, with or without associated bronchial wall and lung parenchymal damage, and classically with pus in the bronchial lumen. There is no knowledge on the prevalence, aetiology, and severity of paediatric bronchiectasis in New Zealand. Primary ciliary dyskinesia (PCD) is an inherited disorder that can cause bronchiectasis and is characterised by specific structural ciliary abnormalities leading to impaired ciliary motility. It has been suggested that ciliary abnormalities may predispose Maori and Pacific Island people to bronchiectasis, but appropriate expertise and non-invasive technology to accurately investigate the condition has not been available in New Zealand. Additionally the exhaled gas nitric oxide (NO), a non-invasive marker of some types of airway inflammation, has been suggested as a useful screening test for PCD. The aims of this thesis were to: 1. Define the demographics, causes, and severity of the known paediatric bronchiectasis population of Auckland. 2. Establish a method for detecting primary and secondary ciliary dysfunction. 3. Explore non-invasive methods for differentiating primary and secondary ciliary disease. 4. Determine the prevalence of PCD in paediatric bronchiectasis in Auckland. Methods: Observations were made on children with bronchiectasis who attended the Starship Children's Hospital, and a cohort of healthy children recruited from local Auckland schools. A retrospective review of the demographics and radiology scores (CXR and HRCT scan) as a measure of disease severity was made. The results were compiled into a bronchiectasis database and a measure of socio-economic factors (NZDep96 index) was incorporated. Equipment was created for the photometric method of assessment of ciliary beat frequency (CBF). After piloting, 3prospective studies were undertaken to evaluate skin prick allergy tests, exhaled and nasal NO, lung function and a nasal brushing for assessment of CBF and ultrastructural analysis in the normal and diseased children. Results. The estimated prevalence of paediatric bronchiectasis in Auckland was ~2/10,000 and was disproportionately more common in the Pacific Island (6.3/10,000) and Maori children (2.8/10,000). Eighty eight percent of cases had bilateral disease, and 64% had 4 or more lobes involved. There was a wide range of presumed aetiologies but over half remained undiagnosed despite extensive investigation. The median duration of symptoms before diagnosis was 3.2 years, and a median of 4 respiratory admissions pre-diagnosis. The NZDep96 index suggested significant associated socio-economic deprivation. A non-invasive protocol to brush nasal epithelium and the technology to assess CBF was created and piloted. Ethnic normal values were established for NO and CBF for healthy European and Pacific Island children. Insufficient Maori children could be recruited. CBF and NO values were not low and comparable with frequencies reported internationally using similar methodologies. Exhaled NO levels did not differ significantly between the children with bronchiectasis and controls, or between the bronchiectatic children who were and were not prescribed inhaled steroids. However CBF and nasal NO were lower in the children with bronchiectasis than controls. The percentage of abnormal ciliary structural defects in the control children was 3 times higher than reported controls, with no difference across ethnic groups. Similar abnormalities were seen in the children with bronchiectasis. These abnormalities were central microtubule defects, tubular additions or deletions, and partial dynein arm defects. In the individual children with bronchiectasis who had low CBF and nasal NO, no single primary ciliary defect was identified to conclusively diagnose PCD. Conclusions: Paediatric bronchiectasis is common and severe in Auckland. New Zealand but the condition has been neglected in terms of recognition. It is hoped that the establishment of a bronchiectasis database for children will not only facilitate collaborative research but also act as a template for a national bronchiectasis database for New Zealand, which can be used to support applications for health resources and funding. Importantly the thesis has resulted in a non-invasive method for assessing ciliary structure and function that could be used to investigate New Zealand children and adults. A wide variety of ciliary abnormalities were found in the New Zealand children that were most likely secondary phenomena, and the incidence of PCD in the population examined, if present, is small. More work is needed to increase the ciliary structural and functional 'library' for New Zealand children, and particularly for Maori children who were under assessed in this work. The possibility of another vulnerability factor, as yet not identified, either of innate immunity or airway defences may still underlie the high prevalence of bronchiectasis in New Zealand.

The amphibian carotid labyrinth was first described in 1738, but its purpose has never been satisfactorily explained. Much of the extant literature has either been restricted to a particular aspect of the organ, or has been based on erroneous information. This thesis used the New Zealand frog Leiopelma hochstetteri because its primitive status might render some characteristics of the carotid labyrinth more clearly evident. The "gross" anatomy of the tiny (0.5 mm diameter) organ was examined by wax-plate reconstruction and by scanning electron microscopy. The cytology of the capillary plexus tissue was studied by light and transmission electron microscopy using conventionally prepared sections. The higher resolution of the electron microscope permitted a more exact study of the innervation of the organ. Neurophysiological experiments were undertaken, but the very short (0.6 mm) and fine carotid nerve proved intractable. A study of the in vivo flow of blood through the labyrinth, resolving the pulsatile variations in flow caused by the action of the heart, was first made by visual observation and then using instrumental methods. The tissues of the capillary plexus contain the same characteristic associations of Type I and Type II cells, together with nerve fibres, as are described in the carotid bodies of the Mammalia, and in the carotid labyrinths of other Amphibia. These associations, which are believed to represent chemoreceptor units, apparently have efferent synaptic complexes against the membrane of the Type I cell. The dense-cored vesicles within the Type I cells fall within the size range described for the other species. Similar characteristic cell groups, complete with nerve supplies, were also observed in the surrounding connective tissue. The carotid nerve contains about 15 myelinated and up to 150 fine (0.15-0.5µm) unmyelinated axons and leaves the sheath of the lingual ramus of the glossopharyngeal nerve through a ganglion of 25-40 cells. It supplies the labyrinth alone. The capillary plexus tissue, in which the characteristic cells are dispersed, is a densely cellular connective tissue and provides a network of random channels for the throughflow of blood. Melanocytes, collagen and elastic fibres, and nerve fibres are embedded in the matrix. The organ is mechanically passive since the only smooth muscle is found as a thin layer in the proximal main chamber. Blood flow measurements utilized the moving erythrocytes as tracers. Their movement, visible through the semi-transparent arterial walls, was recorded with a television camera and later analyzed with a microkymograph. Simultaneous recording of the ECG-related velocity measurements to the phase of the cardiac cycle. Flow in the common carotid was markedly pulsatile, entering the labyrinth in bursts and causing it to swell, while flow in the internal carotid was steady. Flow in the external carotid reversed for a large part of the cycle, returning to the roots of that vessel as the entire organ was distended. The organ owes its pulse-filtering behaviour to the resistive effect of the capillary plexus, combined with a reservoir-like storage due to the compliance of the fibroelastic capsule. This function concurs with Ask-Upmark's observation that all mammals have some pulse-filtering device interposed in their cranial arterial supply. I propose this to be the main function of the labyrinth, accounting adequately for its complex vascular structure. Its undoubted chemoreceptor function clearly can coexist happily with this vascular one.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / Cerebrovascular disease (stroke) is the third leading cause of death in New Zealand and an important cause of chronic disability. An analysis of routinely available national mortality data indicates that death rates have declined significantly since the 1950s and an acceleration of the rate of decline has occurred during the past decade. Reasons for the decline are unknown, although the most probable explanation is a decline in incidence rather than an improvement in case fatality rates. National data however provides little information about the total impact of the disease on a defined population. The Auckland Stroke study was developed to measure the incidence and natural history of the disease, to document the patterns of management, and to describe the range and duration of health services directed at those who survive the acute phase. An innovative sampling method was used to identify a 50% representative sample of all new stroke episodes (WHO criteria) occurring in residents of the Auckland region (total population 829,545) in a one year period; structured interviews were carried out at intervals throughout the year following the onset of the stroke.

This study aimed to determine the relationship between auditory evoked potential measures, speech perception and frequency discrimination in 12 experienced adult cochlear implant (CI) users and to compare the evoked potential results in the CI subjects to those of a group of age- and gender-matched control subjects with normal hearing. The CI subjects all used the Nucleus CI-22 implant with the SPEAK processing strategy. Subject age range was 27-74 years (mean 50.9 years). All had used their implant for a minimum of one year (range 13-76 months, mean duration of implant use 37.5 months).Duration of profound deafness prior to implantation ranged from 1 - 40 years (mean 15.6 years).Middle latency response (MLR), obligatory N1-P2 cortical potentials, mismatch negativity and P3a auditory evoked potential were recorded. Speech perception was evaluated using tests from "The Minimum Speech Test Battery for Adult Cochlear Implant Users". Frequency discrimination was measured using a two-alternative forced choice procedure, to record difference limens (DL) for a 1000Hz tone. Duration of deafness correlated strongly with speech perception scores with poor scores reflecting greater years of deafness. The most sensitive of all the evoked potential measures was P2 latency, with strong relationships found between P2 and duration of deafness and speech perception scores. Earlier P2 latencies were associated with shorter durations of deafness and higher speech perception scores. In general, mismatch negativity was absent or degraded in CI subjects with poor speech scores. In better implant users mismatch negativity was typically present to both large and small frequency differences and demonstrated characteristic changes of increased latency and decreased area and amplitude for the more difficult discrimination task (as seen in the normal hearing group). Na amplitude of the MLR correlated negatively with duration of deafness, with small amplitudes reflecting greater duration of deafness. Frequency DLs were very variable, and no strong relationships were found between these and the evoked potential or speech perception measures.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / Smoking uptake during adolescence continues to present a perplexing public health issue in New Zealand. The plethora of adolescent smoking research is testament to the immense academic effort devoted to developing an understanding of what factors motivate smoking initiation. It is evident that the causal pathway to smoking is both complex and multifaceted and therefore precludes a simple etiological explanation. Favourable portrayals of smoking in popular film have been investigated as a potential motivator to smoking uptake among adolescents. However, few studies have been conducted which explore adolescents’ perceptions of smoking imagery in film or assess how these beliefs and common understandings are moderated by various social demographic and cultural contextual factors. Two principal theoretical approaches, social representation theory and a cultural studies perspective, guided the interpretive analyses of the studies. Three studies were conducted to investigate the role of tobacco imagery in film in the development of an adolescent's self-concept and smoking-related beliefs. The first two studies employed a qualitative research methodology to develop rich, descriptive accounts of young and older adolescents' understandings of tobacco imagery in film. A series of focus group interviews were conducted with Year 8 students (12-13 years) and subsequently with Year 12 students (16-17 years). Groups of 6 - 8 students were randomly selected from a range of Auckland schools. The studies aimed to explore how young people made sense of smoking imagery that are presented in recently viewed films. The two studies (year 8 and Year 12) were conducted independently. During the interviews, participants discussed their recollections of, and responses to, images of tobacco use in recently viewed films. Findings from the studies suggest that younger and older adolescents perceive smoking imagery to be highly pervasive in popular film. In addition, adolescents in the studies were predominantly nonchalant about smoking portrayals in film. However, various factors influenced the extent to which a smoking image was perceived as salient and credible. Images that resonated with expectations and experiences of tobacco use were therefore considered an accurate representation of reality and were more likely to be considered authentic and credible. Within the older adolescent study, findings suggested that images that depicted the lived experience of being a smoker (addiction, cravings, and smoking as stress relief) were especially salient. Alternatively, younger adolescents were more likely to draw upon a scripted knowledge of tobacco use, or familial experiences of being a smoker, to register film representations of smoking as authentic. Evidence of a patently ‘fake’ smoking performance detracted from the salience of an image. Overall, few participants expressed negative responses to the inclusion of smoking imagery in film. A questionnaire was subsequently designed based upon the theoretical constructs extrapolated from the two qualitative studies. A randomised, stratified sampling strategy was used to obtain two representative samples of Auckland Year 8 and Year 12 students (N = 3042). The sample was stratified according to school decile rank, gender and ethnicity. The total sample constituted 48% (n = 1464) Year 8 students and 51.8% (n = 1576) year 12 students. The median age of Year 8 students was 12 years, within the Year 12 group; the median age was 16 years. The questionnaire was designed to assess different perceptions of smoking in film and smoking in real life as a function of age level, gender, smoking status, and ethnicity, on young people's perceptions of smoking imagery in film. The second aim was to assess the extent to which smoking imagery is perceived to be an influential factor in the transition to smoking. The third aim of the study was to develop and test a model that would assess the relationship between the theoretical constructs that emerged from the qualitative studies, and smoking expectations. Findings revealed that age level, gender, smoking status, and ethnicity were associated with perceptions of smoking imagery in film and the normalcy and acceptability of smoking in real life. Specifically, smokers were more likely to perceive that smoking imagery is common in film. A significant difference was also detected between the age levels and smoking status across smoker stereotypes. Age level, smoking status and ethnic group affiliation were associated with perceptions of smoking prevalence among friends and adults. Path analyses were also conducted to test the relationships between the constructs in the theoretical model. Results indicate that perceptions of smoking prevalence, a non-judgmental attitude towards smoking in real life and positive associations with tobacco use in film are associated with smoking in the future. Findings from the studies provide evidence that young people are attentive to smoking imagery in film and these images have differential meanings according to a range of contextual and cultural factors. Smoking imagery in film may play an important supporting role in the construction and maintenance of ambivalent smoking beliefs. Socio-demographic variables, in particular age level and smoking status, are associated with adolescents’ perceptions of smoking imagery in film. Perceptions of smoking in film are also associated with contextual factors such as beliefs about the normalcy and acceptability of smoking in reality. The findings therefore emphasise the relevance of exploring viewers' perceptions of smoking imagery and their lay beliefs about tobacco use when assessing the impact of smoking imagery on smoking outcomes.