Another misleading offering from Icons of Evolution:
VERTEBRATE EMBRYOS. Why do textbooks use drawings of similarities in vertebrate embryos as evidence for their common ancestry -- even though biologists have known for over a century that vertebrate embryos are not most similar in their early stages, and the drawings are faked?
The short answer: most texts no longer do this; editors & publishers have corrected their error as they've become aware of it.
The long answer: in this claim Wells is referring to a series of drawings by the German embryologist Ernst Haeckel, which seemed to show very strong similarities in form between the embryos of a number of different vertebrate groups. Haeckel may well have fudged these drawings to exaggerate the embryos' similarities - but nonetheless, current research does show that early vertebrate embryos do in fact have a number of similarities. And embryos of species that recently shared a common ancestor have more features in common than more distantly-related groups. For example, at some stage in their development all chordate embryos have a notochord, a dorsal nerve cord, a tail that extends beyond the anus, and pharyngeal pouches - features which are missing from all other animal taxa. (These four features are examples of homologous structures.) And within the chordates, mammal embryos are more similar to each other than they are to reptile embryos.
Wells also tries to make out that Haeckel's flawed drawings formed the basis of Darwin's understanding of embryology, which was one line of evidence in The origin of species. In fact, Haeckel's work was published well after Darwin completed The origin, so could hardly have had an impact. What did impress Darwin was Haeckel's work on phylogenetic trees, which supported and illustrated his own key concept of descent with modification. In other words, he was delighted with Haeckel's application of the theory of evolution via natural selection, rather than using Haeckel's work to support his own.
What's more, Wells berates Haeckel, not only for fudging his drawings, but also for not including information on monotremes [echidnas & platypus] in his figures. Now this really is silly. As Alan Gishlick points out, monotremes hadn't even been discovered in 1866, when Haeckel published his famous embryology work! What's more, when the scientific world did get its collective hands on monotreme specimens, monotremes were recognised as mammals because they had common features of embryonic development. This is an excellent example of how important studies of comparative embryology are to our understanding of evolution - and 'evo-devo' (or evolution & developmental biology) is one of the fastest-growing fields in biology today.