In evolution (of the species) there is no reversal
Danilo Anton

The biological evolution develop paths that are marked in the genes of the successive species and that define certain anatomical structures and physiological functions that are printed in all the derived genomes.
For example, when the first tetrapods (four-legged vertebrates) were defined from the ancient Tetrapodomorpha, which in turn evolved from the old finfish-lobes (Sarcopterygii), about 390 million years ago (in the Devonic period) the bone structure of all the vertebrates that succeeded them was defined. 

Thus, batrachians, dinosaurs, various reptiles, birds, mammals, including primates, all have four limbs based on similar structures. These determined the various ways of moving that have all tetrapods: walking, jumping, swimming, crawling, flying,, adapting the limbs for different functions. 
The limbs were transformed into legs, wings, fins or similar organs. In all of them, the articulated components were preserved, allowing the development of their functions.

The conclusion I want to make is that there is no "reverse" in the evolution of vertebrates. In some cases the limbs are atrophied because they are not necessary, they remain as vestiges of the old structures and even disappear (eg the snakes).
Similar situations occur with other components of anatomical structures or physiological functions, such as digestive and circulatory systems, nervous system, etc.
In the development of the human brain the same type of processes occur. There is a "reptilian" brain in the interior, a mammalian brain around and finally a cortex that responds to the most recent anthropic evolution.
In all these cases, it is noticed that the original designs established certain patterns that were conserved as the genomes evolved. But there was no change in basic design or reverse. Only readaptations and aggregates.
One of the main reasons for this to occur in this way is that the evolutionary changes are gradual and that in the successive series of changes each intermediate genome has to give rise to viable individuals. In other words, intermediate genomic individuals must all be viable, capable of competing in the natural environment and therefore capable of reproducing.
An analogy can be made in the development of urban matrices of cities.
Many modern cities developed in times when the transport routes (roads, streets) were adapted to the means of transport corresponding to the time.
Thus, the first roads (for carts, riders, errands) were traced avoiding flood areas (small glens and water courses, marshes) and steep reliefs (ravines, rocky hills, etc.).

When the city grew old wagon roads became streets and avenues. In some cases, its route was modified, but frequently it continued giving rise to streets and avenues with sinuous trajectories. The most modern streets in later developments were designed according to particular local plans, often in the form of checkerboard or other geometric modalities.

This happened because it is difficult to substantially modify the urban basic plan of the cities.
As time passes the radical changes that would be required would be very onerous and there would be social and / or political resistance to carry them out. And so in many cities certain avenues remain with different curvilinear paths to the adjacent streets that are only explained by historical reasons ..
In Montevideo, this is the case of Avenida Agraciada (old carriageway to the west) and 8 de Octubre Avenue - Maldonado Road (carriage road east). The Rivadavia avenue in Buenos Aires was in historical times (century XVIII) the Camino Real towards the west (road of carretas).

In Toronto, Canada (a city where I lived a few years ago) there is a street called Dundas that has similar characteristics showing that in the past it had been a west-east carriageway.
In the cities that were planned since its founding (La Plata in Argentina, Brasilia in Brazil), urban matrices have not undergone major changes, but in practice constitute a small minority in contemporary urban landscapes.