It was ludicrous to suppose that an injury would itself be heritable. Weismann could have done a Thought Experiment that would have been equally convincing. He did the actual experiment to show the Lamarckians in a way that they accepted at the time.
The evolutionary changes that take place in species over time, long long times, would be changes that influence the chance of reproducing, perhaps by a tiny percentage.
Lamarck proposed that a giraffe could lengthen its own neck by stretching and that the longer neck would be a heritable trait. In reality, the evolutionary mechanism would be via reproductive fitness. If a giraffe could lengthen its neck by stretching, it might get more food and produce more babies. Over time, the population might gradually include more of the longer-necked giraffes and slightly fewer of those who just couldn't reach as high.
If it were possible for neck stretching exercises to result in children with a longer neck, a parent who worked out at the gym could expect to have a child with stronger biceps. Well yes, this might in fact happen – but via lifestyle, not via genetic inheritance, if a parent who loves the gym inspires a child to also work their own biceps.
When Weismann removed the tails of his lab mice, he did not influence their survival chances. In the wild, perhaps mice with missing tails would have a lower chance of making it or of having babies. But that would not prove that the injury was inherited, only that the injury influenced the reproductive fitness of the specific injured animals.
