Second, drugs from animal are sometimes ineffective and harmful for humans. Animals’ biology is very different from human beings. The anatomic, metabolic, and cellular of animals are so different from humans’, so the result of the test may be low-qualified, and inaccurate. The differences between animals and humans can lead to the completely different reactions. According to Paul Furlong, Professor of Clinical Neuroimaging at Aston University (UK), "it's very hard to create an animal model that even equates closely to what we're trying to achieve in the human." For example, the Draize eye test−an acute toxicity test−is impractical to human because rabbits’ and humans’ cornea are structurally different. Rabbits’ corneas produce fewer tears, so the chemical will not be cleaned as soon as in humans’ eyes and it will cause more irritation. Another example is skin test. It has been applied to animals in the way that is not appropriate for human, such as via injection, so it has been criticized because animal’s skin is anatomically different from human being’s. In addition, animals are given larger amount of chemicals than human in order to ensure that the chemicals reach the animal’s cell, so the result of the test is impractical for humans because the amount of the chemicals is different. Consequently, different biology also leads to unreliably predicted result. The results are often inaccurate. For example, a study shows that almost 150 clinical treatments failed to reduce the pain in patients, even though they were tested in animal experiments, and 94% of drugs that pass animal tests don’t show the result in the same way as in human.