Search

Background: Diabetic dyslipidemia is characterized by elevated levels of triglyceride-rich lipoproteins and reduced levels of HDL-cholesterol (HDL-C) . Low levels of HDL-C are associated with an increased risk of atherosclerotic cardiovascular disease, because HDL protects against atherosclerosis through multiple mechanisms include removal of excess cholesterol from macrophages. HDL can also influence cholesterol homeostasis in pancreatic β-cell. It is noteworthy that absence of ATP-binding cassette transporter A1 (ABCA1) , which is a cellular cholesterol transporter and regulates islet cholesterol efflux, results in islet cholesterol overload and impaired insulin release in mice. We examined the associations of cholesterol efflux and insulin secretion in MIN6 cells with polyphenols that is reported to increase cholesterol efflux. Methods: Murine MIN6 cells were labeled with 3H-cholesterol and incubated with ApoA1 (μg/ml) or HDL (25 μg/ml) for 24 hours. The percentage cholesterol efflux was calculated by dividing the media-derived radioactivity by the sum of the radioactivity in the media and the cells. The effect of polyphenols on the cholesterol efflux was studied, adding that into the medium before the analysis of the cholesterol efflux. Insulin secretion was assessed by ELISA as the concentration of insulin in the culture medium. The mRNA expression of ABCA1 and ABCG1 were determined by real-time Quantitative RT-PCR. Results: Under the existence of polyphenols (isorhamnetin) , the ApoA1 or HDL-mediated cholesterol efflux had significantly increased by 1.4% or 1.8%, respectively (n=4, p Conclusion: These data suggest that isorhamnetin increases cholesterol efflux and insulin secretion in MIN6 cells. Isorhamnetin could be the interesting candidate to investigate the new treatment for patients with diabetes mellitus. Disclosure K.Matsuki: None. Y.Kimura: None. K.Matsumura: None. J.Tanabe: None. H.Murakami: None. M.Daimon: None. Funding JSPS KAKENHI Grant Number 20K11549

Objective: Type 2 diabetes mellitus (T2DM) is associated with a high prevalence of depression, and depression is influenced by personality traits; however, these psychological factors have not been well studied in individuals with T2DM. The association between the use of a 5-factor model of personality and depressive symptoms was examined in individuals with T2DM. Methods: The subjects were 435 T2DM patients (mean age 63.1±12.6 years). Depressive status and personality traits were assessed using the Center for Epidemiologic Studies Depression Scale and the Ten-Item Personality Inventory, Japanese version, respectively. Lifestyle factors and glycated hemoglobin A1c levels in the patients were also included in the analyses. Results: Among the 435 subjects with T2DM, 130 (29.9%) exhibited possible depression, and 68 (15.6%) exhibited probable depression. After adjustment for confounders, Extraversion, Agreeableness and Neuroticism were found to be significantly associated with the presence of depression. No relationships were found between depression and HbA1c. Conclusion: These findings indicate that Extraversion and Agreeableness are protective factors, and Neuroticism is a risk factor for depression in T2DM patients. Psychological therapy focusing on personality may reduce depressive symptoms. Additional studies are needed to examine the relationships between psychological factors and depressive symptoms using a longitudinal study design.

Objective Glycemic control varies based on lifestyle factors and stress coping mechanisms, which are influenced by personality. The psychological factors associated with glycemic control have not yet been established in patients with type 2 diabetes mellitus (T2DM). The relationship between a 5-factor model of personality and glycemic control was evaluated in individuals with T2DM. Methods The subjects were 503 Japanese outpatients with T2DM. Glycated hemoglobin A1c (HbA1c) levels, depressive status, insomnia and personality traits were assessed. Lifestyle factors of the patients, such as habitual alcohol consumption and smoking, were also included in the analyses. Results Because the influence of insulin therapy on HbA1c is so strong, we stratified the patients according to insulin use. Simple regression analysis showed a significant correlation between HbA1c and neuroticism in patients who did not use insulin. After adjustment for confounders, multiple regression analyses revealed that none of the personality factors, including neuroticism, were found to be associated with HbA1c. Conclusion These findings suggest that personality traits do not have a large impact on glycemic control. Further studies are required to confirm the relationships between psychological factors and glycemic control using a longitudinal study design.

Aims/Objectives Glycemic control varies according to stress level and the efficacy of control measures, affecting the outcomes of diabetes. Although detailed coping styles have not been well studied in patients with type 2 diabetes mellitus, problem‐focused coping strategies are believed to be related to better control of blood glucose. Associations between coping profiles/dimensions and blood glucose control were examined in individuals with type 2 diabetes. Materials and Methods The participants included 503 Japanese patients (mean age 63.9 ± 12.6 years) with type 2 diabetes. The average glycated hemoglobin A1c (HbA1c) levels were calculated from HbA1c levels measured more than four times within the 12 months before the assessment. Coping profiles were assessed using the Brief Scale for Coping Profile. Lifestyle factors were also included in the analyses. Results Factors other than age were not associated with HbA1c levels in patients who used insulin. Conversely, habitual alcohol consumption, single status, the adaptive emotion‐focused coping dimension, and changing mood and changing one’s point of view profiles were associated with HbA1c levels. Conclusions These findings suggest that adaptive emotion‐focused coping supports glycemic control in type 2 diabetes patients who do not use insulin. Additional studies using a longitudinal design are required to further examine the relationships between psychological factors and glycemic control., This study investigated coping styles associated with glucose control in people with type 2 diabetes. An adaptive emotion‐focused coping dimension supports glycemic control in people with type 2 diabetes who do not use insulin.

Objective Type 2 diabetes mellitus (T2DM) is associated with a high prevalence of depression, which is influenced by personality traits and coping style. However, these psychological factors have not been well studied in individuals with T2DM. The association between coping behaviors and the reported levels of depressive symptoms was examined in individuals with T2DM. Methods The subjects were 435 T2DM patients (mean age 63.1 ± 12.6 years). Depressive status, personality traits and coping behaviors were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D) and the Brief Scale for Coping Profile (BSCP). Lifestyle factors and glycated hemoglobin A1c (HbA1c) levels in the patients were also included in the analyses. Results Among the 435 subjects with T2DM, 130 (29.9%) exhibited possible depression, and 68 (15.6%) displayed probable depression. After adjustment for confounders, logistic and multiple regression analyses revealed that certain coping profile scores, such as Changing one’s point of view, Emotional expression involving others and Avoidance and suppression, were consistently and significantly associated with the presence and severity of depression. No relationship was found between depression and HbA1c. Conclusion These findings indicate that Maladaptive emotion-focused coping strategies, such as Emotional expression involving others and Avoidance and suppression, are protective factors and that Adaptive emotion-focused coping, such as Changing one’s point of view, is a risk factor for depression in T2DM patients. Psychological intervention focusing on the coping profile may reduce depressive symptoms. Additional studies are needed to examine the relationships between psychological factors and depressive symptoms using a longitudinal study design.

Objective: Type 2 diabetes mellitus (T2DM) is associated with a high prevalence of depression, and depression is influenced by personality traits; however, these psychological factors have not been well studied in individuals with T2DM. The association between the use of a 5-factor model of personality and depressive symptoms was examined in individuals with T2DM. Methods: The subjects were 435 T2DM patients (mean age 63.1±12.6 years). Depressive status and personality traits were assessed using the Center for Epidemiologic Studies Depression Scale and the Ten-Item Personality Inventory, Japanese version, respectively. Lifestyle factors and glycated hemoglobin A1c levels in the patients were also included in the analyses. Results: Among the 435 subjects with T2DM, 130 (29.9%) exhibited possible depression, and 68 (15.6%) exhibited probable depression. After adjustment for confounders, Extraversion, Agreeableness and Neuroticism were found to be significantly associated with the presence of depression. No relationships were found between depression and HbA1c. Conclusion: These findings indicate that Extraversion and Agreeableness are protective factors, and Neuroticism is a risk factor for depression in T2DM patients. Psychological therapy focusing on personality may reduce depressive symptoms. Additional studies are needed to examine the relationships between psychological factors and depressive symptoms using a longitudinal study design.

Aim: Sterol regulatory element-binding protein (SREBP)-1c is the dominant liver insulin-stimulated isoform and strongly correlates with diabetic dyslipidemia characterized by hyperinsulinemia [i.e., high-density lipoprotein cholesterol (HDL-C) levels and hypertriglyceridemia]. MicroRNA (miRNA) 33b is harbored in the intron of SREBP-1c and represses ATP-binding cassette, sub-family A, and member 1 (ABCA1) expression, essential for HDL formation. We measured plasma miRNA33b levels as possible biomarkers for diabetic dyslipidemia in patients with type 2 diabetes mellitus (T2DM) showing insulin resistance. Methods: The participants included 50 patients with T2DM (M/F 31/19) enrolled in an educational program for controlling blood glucose levels at Hirosaki University Hospital. HbA1c, fasting plasma glucose, insulin, and lipid levels were determined. Plasma miRNA33b, miRNA33a and miRNA148a were quantified using a TaqMan® MicroRNA Assay, and values were corrected with reference to miRNA16. Results: Mean BMI of participants were 28.2 ± 6.6 (kg/m2) and the Homeostasis Model Assessment of Insulin Resistance was 4.3 ± 2.7. Patients' laboratory findings indicated diabetic dyslipidemia with insulin resistance. Plasma miRNA33b/16 levels revealed a positive correlation with plasma insulin level (r = 0.326, P = 0.021), serum C-peptide (r = 0.280, P = 0.049), and triglyceride (r = 0.351, P = 0.012), but no association with HDL-C (r = −0.210, P = 0.143). The blood level of miRNA33a was approximately 1/150th of that of miRNA33b and was not correlated with the above parameters. Conclusion: We postulated that plasma miRNA33b may be useful as a new metabolic biomarker of dyslipidemia in patients with T2DM as well as metabolic syndrome via an insulin/SREBP-1c/miRNA33b/ABCA1 pathway.

Voit and Almeida have proposed the decoupling approach as a method for inferring the S-system models of genetic networks. The decoupling approach defines the inference of a genetic network as a problem requiring the solutions of sets of algebraic equations. The computation can be accomplished in a very short time, as the approach estimates S-system parameters without solving any of the differential equations. Yet the defined algebraic equations are non-linear, which sometimes prevents us from finding reasonable S-system parameters. In this study, we propose a new technique to overcome this drawback of the decoupling approach. This technique transforms the problem of solving each set of algebraic equations into a one-dimensional function optimization problem. The computation can still be accomplished in a relatively short time, as the problem is transformed by solving a linear programming problem. We confirm the effectiveness of the proposed approach through numerical experiments.

This paper proposes a new technique which creates strategy trees for derivative (option) trading investment decisions based on behavioral finance theory and optimizes them by evolutionary computation in order to achieve high profitability. The strategy tree uses technical analysis based on a statistical, experienced technique for investment decisions. The trading model is represented by various technical indexes, and the strategy tree is optimized by genetic programming (GP), a form of evolutionary computation. This paper also proposes a method using the prospect theory based on behavioral finance theory to set the psychological bias for profit and deficit and attempts to select the appropriate strike price of options for higher investment efficiency. This technique was found to produce good results and the effectiveness of this trading model by the optimized dealings strategy was demonstrated. © 2013 Wiley Periodicals, Inc. Electron Comm Jpn, 96(4): 15–28, 2013; Published online in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/ecj.11469

SUMMARY This paper proposes a new optimizing system for stock portfolios that uses evolutionary computation techniques to derive a highly suitable combination and investment ratio of brands as well as an appropriate trading-strategy tree. Accurately predicting price trends in the stock market is a difficult task to achieve with the result that investors often suffer great losses. Because stock portfolios are thought to be a valid means of avoiding such risks in terms of financial engineering, they have the effect of reducing risk by diversifying investment into several different brands. Based on this, it was attempted to determine an optimal combination of brands that constitute a portfolio and to derive the investment ratio using a multiobjective genetic algorithm, and also to optimize a trading strategy tree using genetic programming. When a performance evaluation was carried out, the system was found to generally obtain the operative results by making it possible to obtain stable profits using a combination of low risk brands. The system was also able to realize low risk investments in all test periods.

There are a lot of constructional differences in an English-Japanese phrase template, and that often makes the act of translation difficult. Moreover, there exist various and tremendous phrase templates and sentence to be refered to. It is not easy to prepare the corpus that covers the all. Therefore, it is very significant to generate the translation pattern of the sentence pattern automatically from a viewpoint of the translation success rate and the capacity of the pattern dictionary. Then, for the purpose of realizing the automatic generation of the translation pattern, this paper proposed the new method for the generation of the translation pattern by using the genetic programming technique (GP). The technique tries to generate the translation pattern of various sentences which are not registered in the phrase template dictionary automatically by giving the genetic operation to the parsing tree of a basic pattern. The tree consists of the pair of the English-Japanese sentence generated as the first stage population. The analysis tree data base with 50,100,150,200 pairs was prepared as the first stage population. And this system was applied and executed for an English input of 1,555 sentences. As a result, the analysis tree increases from 200 to 517, and the accuracy rate of the translation pattern has improved from 42.57% to 70.10%. And, 86.71% of the generated translations was successfully done, whose meanings are enough acceptable and understandable. It seemed that this proposal technique became a clue to raise the translation success rate, and to find the possibility of the reduction of the analysis tree data base.

We previously proposed a new way of automatically generating membership functions that play an important role in improving the accuracy of fuzzy control by using genetic programming (GP). In this paper, we report 2 improvements for achieving better fuzzy robot intelligence. First, we made the mutation rate change dynamically based on a coupled chaotic system. Secondly, we introduced population partitioning using deme by parallel processing. The effectiveness of these improvements is demonstrated through computer simulation.

This paper describes a negotiation agent system based on the fuzzy decision making. The method of seeking appropriate membership functions and a reasonable agreement point was examined by means of the genetic programming technique with the coupled chaos system, which is an intelligent principle. The negotiation rule is based on the negotiation model expressed by the utility theory in the process of decision making. And the concession process was modified with the opponent’s movement and the persistence of each negotiator. In order to search for a membership function more efficiently, the dynamic state of symbiosis between individuals, which was caused by the coupled chaos system, was taken advantage of. Then the effectiveness of the technique was examined by applying it to a practical negotiation case which needs cooperative decision making. As a result, the following findings were obtained. This technique helps discover practicable membership functions in a vast search area, and achieve the solution search with high efficiency. This technique is also considered to be applied to the negotiation support easily.

Aims Cholesterol efflux with high-density lipoprotein (HDL) particles has an important role in the first step of reverse cholesterol transport (RCT). However, HDL function in type 2 diabetes has not been well investigated thoroughly. We measured cholesterol efflux in 36 patients with type 2 diabetes compared with 9 controls without diabetes. Methods The HDL fraction was separated with polyacrylamide gel and recovered using the protein recovery system. Concentration adjusted HDL fraction was used to determine HDL-mediated cholesterol efflux (Efflux-hdl) from THP-1 derived macrophages. We measured paraoxonase-1 (PON 1) activity to determine antioxidation capacity, serum amyloid A protein (SAA) to determine inflammatory response, and carboxymethyl-lysin (CML) to determine antiglucoxidative capacity. Results Efflux-hdl demonstrated no correlation with plasma apoprotein A-1 (ApoA-I) or HDL-cholesterol in patients with diabetes. PON1 activity in the patients’ HDL fraction was positively correlated with Efflux-hdl (r = 0.39, p = 0.02), and showed a negative tendency with HbA1c levels (r = −0.28, p = 0.10). SAA and CML levels did not demonstrate correlation with Efflux-hdl in patients with diabetes. Conclusion We confirmed the functional changes in HDL particles in the patients. Efflux-hdl from macrophages was reduced depending upon the decrease in PON1 activity, which was inversely related to HbA1c levels.

For the inference of S-system models of genetic networks, this study proposes a new method, i.e., a two-phase estimation method. The two-phase estimation method is an extension of the decoupling approach proposed by Voit and Almeida. The decoupling approach defines the estimation of S-system parameters as a problem of solving sets of non-linear algebraic equations. Our method first transforms each set of non-linear algebraic equations, that is defined by the decoupling approach, into a set of linear ones. The transformation of the equations is easily accomplished by solving a linear programming problem. The proposed method then estimates S-system parameters by solving the transformed linear equations. As the proposed two-phase estimation method infers an S-system model only by solving linear programming problems and sets of linear algebraic equations, it always provides us with a unique solution. Moreover, its computational cost is very low. Finally, we confirm the effectiveness of the proposed method through numerical experiments.

Practical function optimization problems often contain several constraints. Although evolutionary algorithms (EAs) have been successfully applied to unconstrained real-parameter optimization problems, it is sometimes difficult for these methods even to find feasible solutions in constrained ones. In this study, we thus propose a technique that makes EAs possible to solve function optimization problems with several inequality and a single equality constraints. The proposed technique simply forces individuals newly generated to satisfy the equality constraint. In order to generate these individuals, this study utilizes a Markov chain Monte Carlo (MCMC) method and crossover kernels. While the proposed technique can be applied to any EA, this study applies it to a relatively simple one, UNDX/MGG. Experimental results show that UNDX/MGG with the proposed technique has an ability to solve unimodal and multimodal function optimization problems with constraints. Finally, we show that, although our approach cannot solve function optimization problems with multiple equality constraints, we can convert some of them into those with a single equality constraint.

An S-system model is considered as an ideal model for describing genetic networks. As one of effective techniques for inferring S-system models of genetic networks, the problem decomposition strategy has been proposed. This strategy defines the inference of a genetic network consisting of N genes as N subproblems, each of which is a 2(N+1)-dimensional function optimization problem. When we try to infer large-scale genetic networks consisting of many genes, however, it is not always easy for function optimization algorithms to solve 2(N + 1)-dimensional problems. In this study, we thus propose a new technique that transforms the 2(N + 1)-dimensional S-system parameter estimation problems into (N+2)-dimensional problems. The proposed technique reduces the search dimensions of the problems by solving linear programming problems. The transformed problems are then optimized using evolutionary algorithms. Finally, through numerical experiments on an artificial genetic network inference problem, we show that the proposed dimension reduction approach is more than 3 times faster than the problem decomposition approach.

Recently, we proposed a genetic network inference method using linear programming machines (LPMs). As this method infers genetic networks by solving linear programming problems, its computational time is very short. However, generic networks inferred by the method using the LPMs often contain a large number of false-positive regulations. When we try to apply the inference method to actual problems, we must experimentally validate the inferred regulations. Therefore, it is important to reduce the number of false-positive regulations. To decrease the number of regulations we must validate, this study assigns confidence values to all of the possible regulations. For this purpose, we combine a bootstrap method and the method using the LPMs. Through numerical experiments on artificial genetic network inference problems, we check the effectiveness of assessing the confidence values of the regulations.

Anodization of single-crystalline silicon was carried out under microwave plasma created by electron cyclotron resonance (ECR), and the anodization characteristic and oxide film property were evaluated. The growth rate of the oxide film depended strongly on the oxygen pressure and the growth rate reached the maximum under relatively high oxygen pressure. Since the plasma parameters and substrate temperature were dependent on the oxygen pressure, the oxidation was thought to be dependent on the supply of oxygen atoms from plasma and the generation of silicon atoms and dangling bonds at the Si-oxide interface. The X-ray photoelectron spectra revealed the formation of stoichiometric SiO2. This result provided insight into the oxide growth process. In addition, the electric conduction in an MOS device formed using this oxide was investigated. As a result, it was found that the dc high field conduction prior to the self-healing breakdown was based on the tunneling of electrons injected by Schottky emission through the oxide. Also, the negative resistance observed after the self-healing breakdown was caused by the formation of space charges caused by the impurity ions.

The inference of genetic networks using a model based on a set of differential equations is generally time-consuming. In order to decrease its computational time, we have proposed the inference method using a normalized Gaussian network (NGnet) model. The inferred models however contain many false-positive regulations when we apply the NGnet approach to the genetic network inference problems. This paper proposes the reduced NGnet model and the gradual reduction strategy to overcome the drawbacks of the NGnet approach. Then, in order to verify their effectiveness, we apply the inference method using the proposed techniques to several artificial genetic network inference problems.

A model based on a set of differential equations can effectively capture various dynamics. This type of model is, therefore, ideal for describing genetic networks. The genetic network inference problem based on a set of differential equations is generally defined as a parameter estimation problem. On the basis of this problem definition, several computational methods have been proposed so far. On the other hand, the genetic network inference problem based on a set of differential equations can be also defined as a function approximation problem. For solving the defined function approximation problem, any type of function approximator is available. In this study, on the basis of the latter problem definition, we propose a new method for the inference of genetic networks using a normalized Gaussian network model. As the EM algorithm is available for the learning of the NGnet model, the computational time of the proposed method is much shorter than those of other inference methods. The effectiveness of the proposed inference method is verified through numerical experiments of several artificial genetic network inference problems.

The random number generator is one of the important components of evolutionary algorithms (EAs). Therefore, when we try to solve function optimization problems using EAs, we must carefully choose a good pseudo-random number generator. In EAs, the pseudo-random number generator is often used for creating uniformly distributed individuals. As the low-discrepancy sequences allow us to create individuals more uniformly than the random number sequences, we apply the low-discrepancy sequence generator, instead of the pseudo-random number generator, to EAs in this study. The numerical experiments show that the low-discrepancy sequence generator improves the search performances of EAs.

This paper describes the plasma anodization of aluminum thin films in microwave‐discharge oxygen plasma with electron‐cyclotron‐resonance conditions from the following two points of view. First, the change in the current efficiency with the anodizing conditions and the factors affecting the oxide growth are examined. The current efficiency depends on the microwave input power, the magnetic field, the distance between the specimen and the location where the plasma is generated, the pressure of oxygen gas, and the presence of ozone. Second, some considerations are given on the generation of the atomic oxygen that closely relates to the growth of the oxide film. The generation process was calculated with a simple model and an approximation. The result suggests that atomic oxygen is generated when electrons and positive ions are extinguished by dissociative recombination in the course of ambipolar diffusion in the plasma. In addition, on the supposition that only the rate of atomic oxygen supplied to the surface of the specimen limits the growth rate of the oxide film, the minimum density of atomic oxygen necessary for the oxide growth was estimated. It was found to be maximal under the appropriate conditions of oxygen gas pressure and position of the specimen. However, the atomic oxygen density is also a complicated function of other conditions such as the electron density, the molecular oxygen density, and the electron energy.

In plasma anodization the oxide growth rate decreases with time during constant current anodization. This phenomenon has been studied noting that the surface potential of specimens is closely related to the current efficiency. The oxide growth rate during the plasma anodization of aluminum wires or films is observed to vary depending on the dimensions of specimens and also on the thickness of a wet‐anodic Al2O3 film previously grown on the specimen surface. It turns out that the surface potential changes not only because the plasma‐anodic Al2O3 film itself is not uniform in thickness but also because the surface potential of the oxide film is floating in a plasma. The change of the surface potential was observed using a sounding probe. It is concluded that a difference in the surface potential exists between the central part and the peripheral part of the specimen, and that the difference increases with the increase of the forming voltage. The change of the surface potential may be the cause of the rate‐l...

The enhancement of the growth rate of Al2O3 has been observed during plasma anodization accompanied by optical excitation with low‐power visible argon‐ion laser light. The increase of the current efficiency with light is approximately 20% of the efficiency without it. This enhancement of the oxide growth rate is considered to arise from promoting the generation rate of atomic oxygen, which contributes to the anodization as oxidizing species, in the plasma sheath near the specimen and/or on the specimen surface irradiated by the laser light. Although most of the atomic oxygen is regarded to be generated by the dissociation of molecular oxygens colliding with plasma electrons, the laser excitation might help the cumulative dissociation of molecular oxygens.

The enhancement of the growth rate of Al2O3 has been observed during plasma anodization accompanied by optical excitation with low-power visible argon-ion laser light. The increase of the current efficiency with the laser light is approximately 20% of that without. And this enhancement of the oxide growth rate is considered to arise from promoting the generation rate of atomic oxygens, which contribute to the anodization as oxidizing species, in the plasma sheath near the specimen and/or on the specimen surface irradiated by the laser light. Although most of atomic oxygens are regarded to be generated by the dissociation of molecular oxygens colliding with plasma electrons, the laser excitation might help the cumulative dissociation of molecular oxygens.

The behavior of the species of oxidation dominating the oxide growth in plasma anodization of Al is investigated in this paper. The relationship between the current efficiency and the plasma parameters was observed under the different anodizing conditions with changing the construction of the discharge electrodes and the discharge current in order to clarify the factors contributing to the anodization. The formula of the current efficiency was tentatively derived from a rough model of the growth process of the oxide film with an approximation. The formula, in practice, was consistent with the experimental results. Consequently, it was suggested that the rate‐limiting step in the oxide growth is due to the supply of the atomic oxygens from the plasma to the surface of the specimen and also to their ionization.

By analysis of the current-voltage characteristics of Al-plasma-anodized Al 2 O 3 -plasma systems the high field conduction characteristics of plasma-anodized thin Al 2 O 3 films in the plasma were found to be expressed by the equation J s = J f0 exp(BF) (where J s is the specimen current density, F the electric field strength across the oxide film and J f0 and B are constants); in this experiment B and J f0 were (1.1-1.4)×10 -6 cm V -1 and about 10 -8 A cm -2 respectively. It was ascertained that the current efficiency in the plasma anodization of aluminium becomes a maximum when the surface potential of the specimen is equal to the plasma space potential. From this it was deduced that most of the negative oxygen ions contributing to plasma anodization are generated by the ionization of oxygen atoms adsorbed on the surface of the specimen.